No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Predicting property price index using artificial intelligence techniques: Evidence from Hong Kong
Abstract
Purpose
Booms and bubbles are inevitable in the real estate industry. Loss of profits, bankruptcy and economic slowdown are indicators of the adverse effects of fluctuations in property prices. Models providing a reliable forecast of property prices are vital for mitigating the effects of these variations. Hence, this study aims to investigate the use of artificial intelligence (AI) for the prediction of property price index (PPI).
Design/methodology/approach
Information on the variables that influence property prices was collected from reliable sources in Hong Kong. The data were fitted to an autoregressive integrated moving average (ARIMA), artificial neural network (ANN) and support vector machine (SVM) models. Subsequently, the developed models were used to generate out-of-sample predictions of property prices.
Findings
Based on the prediction evaluation metrics, it was revealed that the ANN model outperformed the SVM and ARIMA models. It was also found that interest rate, unemployment rate and household size are the three most significant variables that could influence the prices of properties in the study area.
Practical implications
The findings of this study provide useful information to stakeholders for policy formation and strategies for real estate investments and sustained growth of the property market.
Originality/value
The application of the SVM model in the prediction of PPI in the study area is lacking. This study evaluates its performance in relation to ANN and ARIMA.
... The results of the study showed the impacts of attributes on property value and predicted satisfactory valuation estimates of the properties. In another comparative study on housing price valuation in Hong Kong, the data related to attributes were collected from a rating and valuation department, �itted into three types of three modelling techniques (ANN, Support Vector Machine (SVM) and Autoregressive Integrated Moving Average (ARIMA)) [13]. In the results, it was shown that ANN predetermined the property value better than the other two modelling techniques. ...
... In the results, it was shown that ANN predetermined the property value better than the other two modelling techniques. The study by [13] lacks information on showing explicit reasons why ANN outperformed the other predicting techniques; however, ANN performed better in terms of capturing nonlinear relationships between input variables and the target variable (property prices), automatically extracting relevant features from the input data, being highly �lexible and adaptable, and managing high-dimensional data and complex relationships more effectively. ...
... Here it is important to note that, in the studies of [6] and [13], the data related to attributes were manually collected from registered real estate �irms and index departments rather than automatically drawn from existing digital databases. ...
Property markets are volatile, necessitating the constant recalculation of property values for a variety of reasons including for the efficient design and construction of buildings. This research is aimed at the automation of the property valuation process using Artificial Intelligence. In a three-part research effort, a Multiple Criteria Decision Method (MCDM) approach using Complex Proportional Assessment (COPRAS) is first applied to predict the property value of new residential units on the basis of a comprehensive list of building characteristic variables identified as relevant for describing a particular property type (in the test case, terraced houses). For initial testing and validation of the valuation prediction calculations, the weights and values of criteria are determined through experts’ opinions and the estimated value of a test property is derived. This first part of the research is described in this report. The second phase of the research involves the automatic acquisition of the variables’ values for any building from the recently digitalised Estonian Building Register. The third part of the research focuses on replacing the need for experts’ opinions of the relative importance weightings of variables through the use of an Artificial Neural Network (ANN) model which is to be trained on existing and continuously refined on new property transaction price data and property characteristics from building permit applications and existing building registers. Parts two and three of this research are still to be carried out and they are outlined in this research paper. It is anticipated that this research will lead to greater efficiency and sustainability through better alignment between building design, construction and market-based property values.
... The real estate market, characterized by its complexity and dynamism, necessitates using advanced predictive tools capable of deciphering the interplay of numerous variables. In recent years, tree-based machine learning models and artificial neural networks (ANN) have emerged as front-runners to enhance prediction accuracy for real estate prices [27][28][29][30][31]. Due to their capacity to capture non-linear relationships and interactions, these models consistently outperform traditional linear models [29,32,33]. However, a significant challenge arises when these models are adopted in real-world scenarios due to their lack of interpretability [34]. ...
... The open data gathered covers a wide range of specific urban indicators, from mobility (bus stops and proximity to subway and train stations) [44,55,56], quality of life and wellbeing (culture, commerce, education, health, leisure, and environment) [35,[57][58][59][60][61], and governance (housing licensing, safety, and security) [21,62,63] to broader macroeconomic and financial indicators (inflation rate, unemployment, gross domestic product, and bank appraisals) [24,33,64,65]. These indicators play a crucial role in influencing the functionality and growth of a smart city and are instrumental in cities' assessment and evaluation. ...
... Following this line of reasoning, the discussion on property prices is enriched by integrating the findings from Abidoye et al. (2019) [33], Vaidynathan et al. (2023) [101], and the SHAP analysis, each contributing unique perspectives on the determinants of real estate valuation. Abidoye's study, focusing on Hong Kong's property market, and Vaidynathan's research on the US housing market underscore the traditional economic view that GDP, CPI, and unemployment rates are key predictors of housing prices. ...
In the rapidly evolving landscape of urban development, where smart cities increasingly rely on artificial intelligence (AI) solutions to address complex challenges, using AI to accurately predict real estate prices becomes a multifaceted and crucial task integral to urban planning and economic development. This paper delves into this endeavor, highlighting the transformative impact of specifically chosen contextual open data and recent advances in eXplainable AI (XAI) to improve the accuracy and transparency of real estate price predictions within smart cities. Focusing on Lisbon’s dynamic housing market from 2018 to 2021, we integrate diverse open data sources into an eXtreme Gradient Boosting (XGBoost) machine learning model optimized with the Optuna hyperparameter framework to enhance its predictive precision. Our initial model achieved a Mean Absolute Error (MAE) of EUR 51,733.88, which was significantly reduced by 8.24% upon incorporating open data features. This substantial improvement underscores open data’s potential to boost real estate price predictions. Additionally, we employed SHapley Additive exPlanations (SHAP) to address the transparency of our model. This approach clarifies the influence of each predictor on price estimates and fosters enhanced accountability and trust in AI-driven real estate analytics. The findings of this study emphasize the role of XAI and the value of open data in enhancing the transparency and efficacy of AI-driven urban development, explicitly demonstrating how they contribute to more accurate and insightful real estate analytics, thereby informing and improving policy decisions for the sustainable development of smart cities.
... The authors Rohmah et al. [14] observed that the Gaussian-RBF outperformed other kernels and found it more suitable for forecasting the CPI . In forecasting Hong Kong's property price index, the authors Abidoye et al. [10] contrasted the ARIMA methodology to two well-known AI approaches: (i) SVM and (ii) Artificial Neural Network (ANN) [10]. They employed the "backpropagation multilayer perceptron ensemble" algorithm to train the ANN. ...
... The authors Rohmah et al. [14] observed that the Gaussian-RBF outperformed other kernels and found it more suitable for forecasting the CPI . In forecasting Hong Kong's property price index, the authors Abidoye et al. [10] contrasted the ARIMA methodology to two well-known AI approaches: (i) SVM and (ii) Artificial Neural Network (ANN) [10]. They employed the "backpropagation multilayer perceptron ensemble" algorithm to train the ANN. ...
... In this work, we employed widely accepted and used TS forecasting approaches to forecast and compared their performances with the proposed MLP. For each WPINonLinear, we developed the following models: Regression [31,32]: Linear (L1), Quadratic (Q), Cubic (C), Logarithmic (L2), and Exponential (E) Exponential smoothing [5,6]: Holt's linear trend (H1), Holt's exponential trend (H2), and Holt-Winters (HW) Auto ARIMA (A) [33,34] SVR [10,14] We used R's stats package to build the regression models [29]. To develop the exponential smoothing and automatic ARIMA models, we employed R software's forecast package [35,36]. ...
Forecasting is an instrument of decision-making that makes predictions or estimates about the future based on historical data. Identifying a suitable strategy for forecasting a time series amongst the classical techniques (e.g., exponential smoothing, ARIMA), Neural approach, and Support Vector Regression (SVR)-another widely used and popular machine learning-based approach, is challenging. The present work aimed at providing a simple (implementation wise), efficient (forecast accuracy wise), and state-of-art Multi-Layer Perceptron (MLP) approach for some selected macroeconomic indices (wholesale price index-i.e., WPI) in India. We looked at the WPIs with non-linear trends identified using the curve-fit method. It's known that the diverse Indian chemical industry contributes notably to India's economic development. In this work, we analyzed the WPI of seventy-seven commodities/items of the "manufacture of chemicals and chemical products" group in India. We detected the indices having non-linear trends by applying the curve-fit method. The curve-fit approach based on statistical rigor identifies the non-linear WPIs. Twenty-five out of seventy-seven indices exhibits non-linear trends. We developed a forecasting approach employing the MLP for these twenty-five non-linear WPIs. The proposed-MLP optimized by hyperparameter tuning offers high accuracy, prediction reliability, and prediction acceptability for all non-linear WPIs. The forecasting performances of the proposed-MLP compared with regression models (Linear, Quadratic, Cubic, Logarithmic, Exponential), exponential smoothing (Holt linear trend, Holt exponential trend, Holt-Winters), state-of-art Auto-ARIMA, and SVR. The MLP outperformed them all. In terms of MAPE, the MLP beat Linear in 88%, Quadratic in 92%, Cubic in 88%, Logarithmic in 72%, exponential in 88%, Holt Linear in 80%, Holt Exponential in 76%, Holt-Winters in 72%, Auto-ARIMA in 56%, and SVR in 56% of cases. We suggest the application of the proposed approach as an alternative for forecasting these twenty-five non-linear WPIs.
... More importantly, they ignore less popular yet complex models. To address this shortcoming, advanced ML approaches have been introduced, including artificial neural networks (ANN) (Abidoye et al., 2019;Xu & Zhang, 2023;Khrais & Shidwan, 2023;Xu & Zhang, 2024), random forests (Park & Bae, 2015;Levantesi & Piscopo, 2020;Soltani et al., 2022;Adetunji et al., 2022;Rey-Blanco et al., 2024), and deep learning models (Yu et al., 2018). Many studies, including Truong et al. (2020) and Manasa et al. (2020), present strong evidence that these advanced ML models outperform traditional approaches. ...
... Scholars have increasingly utilized ML methods to estimate property prices, continually showing improved prediction abilities (Zulkifley et al., 2020;Thamarai & Nakarvizhi, 2020;Begum et al., 2022). The methods discussed in this study include ANN (Nguyen & Cripss, 2001;Abidoye et al., 2019), random forests (Park & Bae, 2015;Levantesi & Piscopo, 2020), and deep learning models (Yu et al., 2018). Empirical research suggests that ML models can harness the advantages of many methods, resulting in enhanced predictive accuracy (Truong et al., 2020;Manasa et al., 2020). ...
Predicting housing prices is particularly of interest to many scholars and policymakers. However, housing prices are highly volatile and difficult to predict. This study used both traditional and advanced machine learning (ML) approaches to address the issue of housing price prediction. This study involves and compares the predictive power between advanced ML models, including random forest, gradient boosting, k-nearest neighbors (KNN), bagged classification and regression trees (CART), and traditional ML models based on linear regression and its modifications. Notably, in this study, we employed both performance metrics, including the mean absolute error (MAE), root mean square error (RMSE), coefficient of determination (R2), and k-fold cross-validation (CV) procedure in order to investigate the predictive performance of each model. Empirically, based on a dataset comprising 78,704 real estate sales in Hanoi, Vietnam, we find that advanced ML approaches outperform traditional approaches. Specifically, advanced ML models enhance the accuracy of house price prediction and the decision-making process related to housing buying and selling activities. Our findings also reveal that among advanced ML algorithms, the random forest algorithm performs better than the other models in predicting housing prices.
... The use of advanced analytical techniques has become increasingly important in understanding and predicting property prices. Abidoye et al. (2019) demonstrated that artificial intelligence (AI) techniques, such as artificial neural networks (ANN), outperform traditional models like ARIMA in predicting property price indices, providing more reliable forecasts [14]. Similarly, Ho et al. (2021) found that machine learning algorithms, particularly Random Forest (RF) and Gradient Boosting Machine (GBM), significantly enhance the accuracy of property price predictions compared to traditional methods [15]. ...
... The use of advanced analytical techniques has become increasingly important in understanding and predicting property prices. Abidoye et al. (2019) demonstrated that artificial intelligence (AI) techniques, such as artificial neural networks (ANN), outperform traditional models like ARIMA in predicting property price indices, providing more reliable forecasts [14]. Similarly, Ho et al. (2021) found that machine learning algorithms, particularly Random Forest (RF) and Gradient Boosting Machine (GBM), significantly enhance the accuracy of property price predictions compared to traditional methods [15]. ...
This paper presents an analysis of property prices in the virtual world, focusing on geographical distribution and district comparisons. Utilizing a dataset of virtual properties, we applied scatter plot analysis, cluster analysis using DBSCAN, and box plot comparison to identify key patterns and opportunities within this market. The scatter plot analysis revealed that property prices are unevenly distributed, with higher prices clustering in specific regions, indicating areas of higher desirability and value. The DBSCAN clustering identified distinct high-value clusters, each containing 10 to 67 properties, and highlighted 1,067 properties as noise, suggesting a dispersed distribution of lower-value properties. Box plot comparisons across districts showed significant variations in property values. Some districts exhibited higher median prices, with the highest at 35,452.60 MANA, while others had lower medians. Variability within districts varied, with some showing a wide range of prices and others more uniform values. Outliers suggested unique investment opportunities in both premium and undervalued properties. For virtual real estate investors, the findings emphasize the importance of location and strategic investment. High-value districts and emerging areas offer potential for significant returns. Developers and urban planners can use these insights to focus on high-demand areas, enhancing project value through strategic investments in infrastructure and amenities. This study highlights the dynamic nature of the virtual real estate market and the importance of ongoing research to understand factors influencing property values. Stakeholders can make informed decisions and capitalize on opportunities in this evolving market.
... Non è tuttavia possibile assegnare alle macchine a vettori di supporto il ruolo di miglior algoritmo per la valutazione in assoluto: tutto dipende dalla natura dei dati a disposizione. Ad esempio, nel loro confronto con le reti neurali artificiali alcuni autori le individuano come più efficaci (Lam, Yu e Lam, 2009), altri autori come meno efficaci (Abidoye et al., 2019;Phan, 2018). Nonostante il crescente interesse che la ricerca sta dimostrante verso tali algoritmi, alle macchine a vettori di supporto nella valutazione immobiliare la letteratura non ha ancora dedicato un dibattuto strutturato come quello che hanno conosciuto le reti neurali artificiali. ...
... However, it is not possible to assign to the machines with support vectors the role of best algorithm for the absolute evaluation, everything depends on the nature of the available data. For example, in their comparison with artificial neural networks, some authors identify them as more effective(Lam, Yu and Lam, 2009), while others identify them as less effective(Abidoye et al., 2019;Phan, 2018). ...
... (Valier 2020) finds that machine learning models perform better than traditional regression analysis in automated valuation when it comes to real estate value estimates. (Abidoye et al. 2019) investigate neural network, support vector machine, and autoregressive integrated moving average models in order to anticipate the Hong Kong property price index. When compared to the other two models, they find that the artificial neural network model works better. ...
The Chinese real estate market has expanded at such a quick rate over the last two decades, up to the current decline patterns that began at the end of 2021. As a result, predicting future property prices has become a significant challenge for both the government and investors. Within the scope of this investigation, we investigate quarterly national residential property price indices for China with data sourced from Bank for International Settlements from the second quarter of 2005 to the first quarter of 2024 by using Gaussian process regressions with a variety of kernels and basis functions. For the purpose of model training and conducting forecasting exercises using the estimated models, we make utilisation of cross-validation and Bayesian optimisations based upon the expected improvement per second plus algorithm. Use of Bayesian optimisations could help endow Gaussian process regression models with good flexibility for forecasting into the future. With a relative root mean square error of 0.1291 percent, root mean square error of 0.1816, mean absolute error of 0.1527, and correlation coefficient of 99.901%, the created models were able to reliably anticipate the price indices from the third quarter of 2020 to the first quarter of 2024 out of sample. The constructed Gaussian process regression models also outperform several alternative machine learning models and econometric models. Their forecast performance is robust to different out-of-sample evaluation periods as well. In order to build hypotheses about trends in the residential real estate price index and to carry out more policy research, our findings might be used either alone or in combination with other projections.
... The literature review indicates gaps in the use of machine learning (ML) for real estate valuation. While many studies highlight the benefits of specific ML techniques in various settings, there is a lack of analyses that compare different ML methods, with Tchuente and Nyawa [27] and Abidoye et al. [37] as exceptions. This makes it harder to generalise the findings and determine the best approaches. ...
The housing market in Dar es Salaam, Tanzania, is expanding and with it a need for increased market transparency to guide investors and other stakeholders. The objective of this paper is to evaluate machine learning (ML) methods to appraise real estate in formal and informal housing markets in this nascent market sector. Various advanced ML models are applied with the aim of improving property value estimates in a market with limited access to information. The dataset used included detailed property characteristics and transaction data from both market types. Regression, decision trees, neural networks, and ensemble methods were employed to refine property appraisals across these settings. The findings indicate significant differences between formal and informal market valuations, demonstrating ML’s effectiveness in handling limited data and complex market dynamics. These results emphasise the potential of ML techniques in emerging markets where traditional valuation methods often fail due to the scarcity of transaction data.
... To improve the risk management process, it is essential to devise effective methods and tools that can predict the potential outcomes of treatment strategies, leading to a more robust and comprehensive approach. Artificial intelligence (AI) manifests itself as a useful methodology to monitor the impact of strategy variations on the predicted variables [27]. For example, neural networks are a powerful tool in terms of risk monitoring [28,29]. ...
This study introduces an integrated method for managing process risks in a business process reengineering (BPR) project using robust data envelopment analysis (RDEA) and machine learning (ML). The goal is to prioritize risks based on three standard factors of PFMEA (severity, occurrence and detection (S-O-D)) and incorporating two additional factors (breakdown cost and breakdown duration) seen as undesirable outputs. The model also accounts for the effect of uncertainty on expert-estimated values by applying disturbance percentages in the linear PFMEA-RDEA model. A machine-learning model is proposed to predict new values if partial or total modifications have been made to the processes. The approach was implemented in an automotive sector company, and the results showed the impact of uncertainty on values by comparing different approaches, such as RPN, PFMEA-DEA and PFMEA-RDEA. A new reduced risk categorization was achieved, which allowed for decision makers to focus on the necessary actions for reengineering.
... AI (Artificial Intelligence) manifests itself like a useful methodology to monitor the impact of strategy variations on predicted variables. [27] For example, neural networks are a powerful tool in terms of risk monitoring. [28,29] There are two key advantages to using neural networks for estimation or prediction tasks. ...
This study introduces an integrated method for managing process risks in a Business Process Reengineering (BPR) project, using Robust Data Envelopment Analysis (RDEA) and machine learning (ML). The goal is to prioritize risks based on three standard factors of PFMEA: Severity, Occurrence, and Detection (S-O-D), and incorporating two additional factors (Breakdown Cost and Breakdown Duration) seen as undesirable outputs. The model also accounts for the effect of uncertainty on expert-estimated values by applying disturbance percentages in the linear PFMEA-RDEA model. A machine learning model is proposed to predict new values if partial or total modifications have been made to the processes. The approach was implemented in an au-tomotive sector company, and the results showed the impact of uncertainty on values by com-paring different approaches such as RPN, PFMEA-DEA, and PFMEA-RDEA. A new reduced risk categorization was achieved, who allowed for decision-makers to focus on necessary actions for reengineering.
... The results show that H2 is supported with a standardized coefficient, b = 0.28 at a significance of p 0.014. This result confirms the results of the previous Abidoye et al. (2019) investigated the use of AI to predict property price indexes. The study also demonstrates the impact of Olarewaju et al. (2020) developed a framework to understand the applicability of an AI-based model in predicting stock prices in the Nigerian market. ...
Artificial intelligence (AI) is digitalizing transportation at sea, on land, and in the air. It has
the potential to cut human error and make operations faster. However, AI is one part of a
broader process to digitize and improve port operations. AI digitalization and autonomous
shipping are critical in the port world; AI allows human work to be shifted toward digital
platforms that are currently not fully capable. Following this, the ports industry can be
sketched as a natural fit for applying AI technology, known for its complicated processes and
the high proportion of human work intervention. This paper aims to analyze and explore the
artificial intelligence Business Drivers of Smart Port Digitalization. This study employs the
exploratory Factor Analysis (EFA), Confirmatory factor analysis (CFA), and structural
equation modeling (SEM) approaches using Advance Managed Outsourced Solutions (AMOS)
based on a ports communities operation management view. Then propose new AI-Ports
initiatives. The value-added core tasks of ports are examined to determine the possible
utilization of AI technology and the AI adoption within the ports community
... The results show that H2 is supported with a standardized coefficient, b = 0.28 at a significance of p 0.014. This result confirms the results of the previous Abidoye et al. (2019) investigated the use of AI to predict property price indexes. The study also demonstrates the impact of Olarewaju et al. (2020) developed a framework to understand the applicability of an AI-based model in predicting stock prices in the Nigerian market. ...
Artificial intelligence (AI) is digitalizing transportation at sea, on land, and in the air. It has the potential to cut human error and make operations faster. However, AI is one part of a broader process to digitize and improve port operations. AI digitalization and autonomous shipping are critical in the port world; AI allows human work to be shifted toward digital platforms that are currently not fully capable. Following this, the ports industry can be sketched as a natural fit for applying AI technology, known for its complicated processes and the high proportion of human work intervention. This paper aims to analyze and explore the artificial intelligence Business Drivers of Smart Port Digitalization. This study employs the exploratory Factor Analysis (EFA), Confirmatory factor analysis (CFA), and structural equation modeling (SEM) approaches using Advance Managed Outsourced Solutions (AMOS) based on a ports communities operation management view. Then propose new AI-Ports initiatives. The value-added core tasks of ports are examined to determine the possible utilization of AI technology and the AI adoption within the ports community.
... Energy is a very popular domain, where the definition of households is carried out for various purposes, such as detecting household building energy anomalies, which might cause cost changes in the energy monthly bills (Himeur et al. 2021) and understanding how households make energy consumption decisions within a technological and institutional context, e.g., adoption of energy-efficient appliances has led to a gradual reduction in household energy use over time (Burnett and Kiesling 2022). Moreover, the study in (Abidoye et al. 2019) uses the household size as one of the most significant variables that can be used for influencing the property price in Hong Kong. Such information can also be used by the government for property-price control to make properties more affordable. ...
Several public services in Malta operate under the stewardship of different governmental bodies, ministries, or departments. This results in considerable effort in the delivery of public services, especially ones that require the use of multiple registries, such as integrated public services (IPSs). Co-creation and co-production are increasingly being seen by public administrations as an approach toward mitigating issues stemming from such a siloed environment. Indeed, they are seen as a means to improve service provision through the delivery of citizen-centric public services that are more efficient and effective. This paper presents the Malta pilot as part of the inGOV project. The latter aims to develop and deploy a comprehensive IPS holistic framework and ICT mobile tools that will support IPS co-creation and governance. The Malta pilot focuses on modernising the Digital Common Household Unit public service. Improving considerably upon the previous ad hoc solution, the Digital Common Household Unit public service implements an iterative co-creation and co-production approach with the various stakeholders. This paper therefore presents the applied methodology in researching current challenges and enablers to the co-creation and co-production of a digital common household unit public service, with a specific focus on sustainability.
... The power to measure the real estate value of the above equation obtained from the C-DHM, which was created to determine the value of the real estate, was found to be at the rate of R 2 0.85. This ratio shows that the C-DHM predicts the market value with high success according to most of the calculations made with real estate value estimation model approaches (Yakub et al., 2020;Yildirim, 2019;Abidoye et al., 2019). Accordingly, in performance calculations C-DHM was applied to the dataset of the study area. ...
In this study, for mass real estate appraisal forecasting, the hybrid mathematical model has been developed by combining Cobb-Douglas one of the nonlinear regression models, and linear modeling. The real estate attributes that create the model were grouped under four main-title: local, spatial, physical and legal features. While Cobb-Douglas was used for the value forecast based on the real estate attributes in each part of the model, an integrated model was created with a linear approach. As a different approach, local and spatial features, which are among the real estate attributes, were used as indexes for reasons such as preventing data confusion in the model and using according to the spatial analysis results of distances. Local and spatial index were prepared with the Fuzzy Analytic Hierarchy Process (FAHP) method to use within the model. For indexes, in the central districts of Konya, 10 local-specific attributes were used, while 12 spatial-specific attributes. The data set has been prepared using legal and physical attributes with market values collected from 457 parcels in the study area. Local and spatial attributes were added as indexes to the data set used in the hybrid model. In addition, modeling was done with the data set used in the Cobb-Douglas Hybrid Model (C-DHM) according to the Linear Multiple Regression Analysis (Linear MRA) method. The developed C-DHM’s results was integrated with Geographical Information Systems (GIS). The performance values between the hybrid model and market values were examined. Results showed that R2 value for C-DHM and Linear MRA used as indexes was found to be 0,85 and 0,80. When the values obtained from C-DHM and market value are compared, it is seen that model gives successful results.
... Various neural network models are used for price forecasting in the market using different indexes, such as the New York Stock Exchange (NYSE) [45], NASDAQ stock exchange [46], price of gold stock in the NYSE [47], real estate price in Hong Kong [48], and Apple stock in NASDAQ [49]. ...
The selling price is one of the essential variables in decision making for fishers regarding the catching of a fishing resource. In the case of the Pacific Mexican lobster fishery, the price uncertainty at the beginning of the season translates into the suboptimal utilization of this resource. This work aims to predict the export price of Mexican red lobster (Panulirus) in a fishing season using demand-related market variables including price, main competitors, main buyers, and product quantities exported/imported in the market. We used the monthly export price from 2006 to 2018 for the main importer, China. As a method for price forecasting, artificial neural networks (ANNs), with and without exogenous variables (NARX, NAR), were used as an autoregressive model, while the same information was analyzed with an ARIMAX model for comparative purposes. It was found that ANNs are a useful tool that yielded better predictive power when forecasting Mexican lobster export prices compared to ARIMAX models. The predictive power was evaluated by comparing the mean square errors (MSE) of 15 models. The MSE of ANNs (73.07) was lower than that of the four ARIMAX models (88.1). It is concluded that neural networks are a valuable tool for accurately predicting prices relative to real values, an aspect of great interest for application in fishery resource management.
... The ANN performs a higher R square and lower Mean Standard Error (MSE) than the hedonic pricing model (Selim, 2009), which indicates that ANN is an effective way to predict housing price. Also, the ANN model outperformed the Support Vector Machine and ARIMA models (Abidoye et al., 2019). ...
This book covers artificial intelligence and ubiquitous cities. It discusses the applications of the relevant tools in bringing revolutionary new lives to mankind. It showcases various applications of artificial intelligence in benefiting human society. For example, AI classification shortens the human time required for classifying court cases; humanoid robots help us perform heavy-duty jobs like humans, connect all the smart home devices, and take care of the kids and the elderly. It also presents the application of AutoML to predict housing prices.
... In addition, some scholars have used other models of machine learning to predict housing prices in cities in the Guangdong-Hong Kong-Macao Greater Bay Area, such as Hong Kong, Macau, Shenzhen, and Guangzhou. Abidoye et al. collected variables affecting house prices in Hong Kong, such as interest rate, unemployment rate, and family size, and fitted them with ARIMA, ANN, and SVM models to generate forecasts of real estate prices [7]. Fong and Wah gathered multiattribute datasets from the Macao SAR Government's Statistics and Census Service and used several data mining methods and algorithms such as SVM, Neural Network, C&R Tree, Weka, SPSS, and Multilayer Perception model to forecast Macao home prices [8]. ...
Situated in southern China, Zhaoqing City is a part of Guangdong Province, China. The total administrative area of the city covers 14,891 square kilometers. The data of China’s seventh population census in 2020 showed that the permanent resident population in Zhaoqing City reached up to 4,413,594. Meanwhile, Zhaoqing is one of the cities in the Guangdong-Hong Kong-Macao Greater Bay Area. House price analysis and prediction carried out against Zhaoqing City will have directive significance for relevant policies formulated by the local government, residential investment or purchase of consumers, and prediction of house price trend as well as business decisions made by enterprises. By virtue of machine learning and statistical theory, the house price in Zhaoqing City from 2010 to 2020 will be researched, and the house price prediction model of Zhaoqing City will be constructed in this paper with several variables including GDP, proportion of tertiary industry, income of urban residents, fiscal revenue, land price, investment volume in real estate development, permanent resident population, population density, and proportion of urban population in net migration. First of all, the methods of correlation analysis will be utilized, to select variables that are highly correlated with house price data based on correlation coefficients. Then, the model will be constructed for predicting the house price on the basis of multiple linear regression analysis that is conducted with selected variables. Finally, the prediction model will be adjusted gradually based on data with different correlations selected from available data, to realize better imitative effect and more precise predictive effect and select optimum prediction model. By means of the above model, the house prices of Zhaoqing City in 2021 and beyond will be predicted accurately, with preferable fitting effect and prediction effect.
... Likewise, Abidoye et al. (2019) investigated the role of AI in forecasting property price index by using an artificial neural network, autoregressive integrated moving average (ARIMA), and support vector machine model. Ayoade et al. (2019) evaluated the implications of financial interoperability linked with new consumers in property development and identified the significance of the "Community Land Trust Shared Equity Housing Model". ...
The Covid-19 pandemic outbreak across the globe has disrupted human life and industry. The pandemic has affected every sector, with the real estate sector facing particular challenges. During the pandemic, property management became a crucial task and property managers were challenged to control risks and disruptions faced by their organizations. Recent innovative technologies, including artificial intelligence (AI), have supported many sectors through sudden disruptions; this study was performed to examine the role of AI in the real estate and property management (PM) sectors. For this purpose, a systematic literature review was conducted using structural topic modeling and bibliometric analysis. Using appropriate keywords, the researchers found 175 articles on AI and PM research from 1980 to 2021 in the SCOPUS database. A bibliometric analysis was performed to identify research trends. Structural topic modeling identified ten emerging thematic topics in AI and PM. A comprehensive framework is proposed, and future research directions discussed.
... The development of effective methods and tools to predict the potential outcomes of treatment strategies is important to enhance the risk management process towards a more holistic and robust approach. Artificial Intelligence (AI) serves as a useful methodology for monitoring the influence of variations in strategy pertaining to the predicted variables (i.e., scenario prediction) [14]. As an example, neural networks have been shown as an promising tool for periodic risk monitoring purposes [15,16]. ...
An integrated method comprising Data Envelopment Analysis (DEA) and machine learning (ML) for risk management is proposed in this paper. Initially, in the process of risk assessment, the DEA cross-efficiency method is used to evaluate a set of risk factors obtained from the Failure Mode and Effect Analysis (FMEA). This FMEA-DEA cross-efficiency method not only overcomes some drawbacks of FMEA, but also eliminates several limitations of DEA to offer a high discrimination capability of decision units. For risk treatment and monitoring processes, an ML mechanism is utilized to predict the degree of remaining risk depending on simulated data corresponding to the risk treatment scenario. Prediction using ML is more accurate since the predictive power of this model is better than that of DEA which potentially contains errors. Based on a case study with a group of logistics service providers, the results ascertain that the combined DEA and ML approach offers a flexible and reasonable alternative in risk management. The approach allow decision-makers or managers to assess and monitor the risk level for handling forthcoming events in unusual conditions. It also serves as a useful knowledge repository such that appropriate risk mitigation strategies can be planned, along with the predicted results. The outcome of our empirical evaluation indicates that the proposed approach contributes towards robustness in sustainable business operations.
... is analysis demonstrates that systems of mass valuation and real estate taxation are an important and viable basis for increasing government revenues. e authors of [1,[11][12][13][14] also note that high-precision methods of real estate valuation are a useful decision-making tool in the taxation and urban planning sectors. Such methods can be used by investors, buyers, and governments. ...
In the modern scientific literature, there are many reports about the successful application of neural network technologies for solving complex applied problems, in particular, for modeling the urban real estate market. There are neural network models that can perform mass assessment of real estate objects taking into account their construction and operational characteristics. However, these models are static because they do not take into account the changing economic situation over time. Therefore, they quickly become outdated and need frequent updates. In addition, if they are designed for a specific city, they are not suitable for other cities. On the other hand, there are several dynamic models taking into account the overall state of the economy and designed to predict and study the overall price situation in real estate markets. Such dynamic models are not intended for mass real estate appraisals. The aim of this article is to develop a methodology and create a complex model that has the properties of both static and dynamic models. Moreover, our comprehensive model should be suitable for evaluating real estate in many cities at once. This aim is achieved since our model is based on a neural network trained on examples considering both construction and operational characteristics, as well as geographical and environmental characteristics, along with time-changing macroeconomic parameters that describe the economic state of a specific region, country, and the world. A set of examples for training and testing the neural network were formed on the basis of statistical data of real estate markets in a number of Russian cities for the period from 2006 to 2020. Thus, many examples included the data relating to the periods of the economic calm for Russia, along with the periods of crisis, recovery, and growth of the Russian and global economy. Due to this, the model remains relevant with the changes of the international economic situation and it takes into account the specifics of regions. The model proved to be suitable for solving the following tasks: industrial economic analysis, company strategic and operational management, analytical and consulting support of investment, and construction activities of professional market participants. The model can also be used by government agencies authorized to conduct public cadastral assessment for calculating property taxes.
1. Introduction
The authors of many recent publications, for example [1], emphasize that artificial neural networks (ANN) as complex nonlinear systems can take into account an unlimited number of external factors and dynamic interactions. Due to this, an ANN allows for solving multiple complex real problems that could not be solved by other methods. For example, the authors of this article created the world’s first neural network lie detector [2]. The ANN technology ensured a unique diagnostic accuracy of 98 percent due to a comprehensive accounting of psychophysiological parameters of an individual such as personal data and signals coming from the sensors of a polygraph machine. The members of the same author’s team developed a neural network system that helps detectives to investigate crimes, in particular, to identify people who can be serial maniacs and murderers [3]. This objective cannot be solved by other methods due to the need to analyze a large number of parameters and factors while many of them have an insignificant impact on the diagnostic result. The same authors created a medical system based on neural network technologies [4–6]. Due to a complex mathematical formulation of the problem, this system allows not only for diagnosing cardiovascular diseases, but also for making disease development forecasts for many years to come and for selecting the optimal courses of disease treatment and prevention (https://en.kardionet.ru). One more neural network system developed by the same team has a practical value enabling users to predict the future box office of a movie based on a set of various factors that have both a direct and an indirect impact on the result of forecasting [7]. Equally important in practice is a neural network system that predicts the bank failure probability and allows you to develop recommendations for preventing such bankruptcies [8]. The book in [9] by the author of this paper provides examples of neural network intelligent systems developed under his leadership and designed to diagnose complex technical devices, the economic position of enterprises, to predict political events, to identify the business and research skills of individuals, etc. Due to their complex formulation of mathematical problems, all these neural network systems enable users not only to diagnose and predict, but also to explore the simulated domains, as well as developing measures for active management of the behavior of these rather complex areas.
As noted by the authors in [1], the cutting-edge capabilities of neural networks make it possible to successfully apply them for modeling complex multifactor nonlinear systems such as a real estate system.
Many authors draw the attention to the great urgency of developing high-precision models for carrying out the mass valuation of real estate markets. For example, [10] presents the results of an analysis of international literature and interviews with statesmen of many countries. This analysis demonstrates that systems of mass valuation and real estate taxation are an important and viable basis for increasing government revenues. The authors of [1, 11–14] also note that high-precision methods of real estate valuation are a useful decision-making tool in the taxation and urban planning sectors. Such methods can be used by investors, buyers, and governments.
As noted in [15], until 1990, five standard recognized methods were mainly used for evaluating real estate such as the comparative method (comparison), contractor’s method (cost method), residual method (development method), profits method (accounts method), and investment method (capitalization/income method).
In the 1990s, some researchers reported about successful attempts to create systems for mass appraisal of real estate objects based on a new mathematical apparatus, artificial neural networks (ANN). Apparently, one of the first papers in this direction was an article [16] published in 1991. Its author, Borst, defined a number of variables for designing an ANN-based model for evaluating New York real estate. He reported that the model can predict the price of real estate with an accuracy of up to 90%. It was a perceptron-type neural network.
In 1991, Tay and Ho in [17] reported on the use of a multilayer perceptron to determine the market price of real estate in Taiwan.
In the same year of 1991, Evans, James, and Collins in [18] reported on the use of neural networks for evaluating residential real estate in England and Wales. After testing several methods, the authors came to the conclusion that the neural network model is best suited for delivering real estate valuations.
In 1992, Do and Grudnitskiy [19] published a report on using a perceptron-type neural network to evaluate US real estate. Based on a test set of 105 houses, the neural network model had twice the accuracy of the predicted values as compared to the analogous regression model.
From the mid-1990s to the present, a series of research publications devoted to the development and application of neural network models for mass appraisal of real estate objects have been published. Many papers [12, 20–32] emphasize the advantages of this advanced technology as compared to regression modeling and other methods of real estate valuation.
Analyzing the papers devoted to neural network modeling of estate markets, it can be noted that few researchers (e.g., [33]) have paid attention to the specific problems of modeling this subject area and to the issues of overcoming these problems. When constructing a neural network system for assessing real estate, the authors in [33] faced the challenge of overcoming the negative impact of statistical outliers on the accuracy of the created models. For the real estate market, they tested a number of methods for detecting outliers such as Tukey’s method, standard deviation method, median method, Z-score method, MAD method, and modified Z-score method. As a result, they concluded that the median method delivers the best results.
Looking ahead, we note that in our work we used an even more effective author’s method for detecting statistical outliers [34] based on the neural network mathematical apparatus.
Summarizing the review of neural network models designed for mass real estate valuation [10, 12, 16–32], let us pay attention to their overall disadvantages:(1)Developed for a specific city, these models cannot be applied to other cities because they do not take into account mesoeconomic factors.(2)All these models quickly become outdated and require frequent updates because they do not take into account the changing economic situation in the world, some specific country, and region over time. Such models can be called static ones. This disadvantage of static models is particularly relevant for developing countries where markets are in the process of development. These markets depend on time-varying oil prices, the dollar, GDP, stock indexes, government credit policies, and so on.
It should still be noted that there is a series of research papers, for example [1, 35], devoted to the development of economic and mathematical models of real estate markets that consider many macroeconomic parameters. However, these dynamic models are intended exclusively for modeling and studying market dynamics. They are not intended for the mass assessment of apartment prices that have a large variety of static characteristics. The apartment cost indices calculated in such models (the average unit cost of apartments assigned to a square meter) can, of course, be recalculated in the cost of specific apartments taking into account their construction, operational, environmental, and other parameters. However, such a recalculation can only be made using additional methods which are not used for mass appraisal of real estate objects due to their inefficiency. The fact is that the unit prices of apartments of the same type located in the same area and even in the same house may differ. Therefore, a more differentiated approach is required in this case.
Thus, on the one hand, we have a list of static models [10, 12, 16–32], etc., for mass appraisal of real estate objects. However, these models do not take into account the changing economic situation in the world, in the country, and in the region over time. Therefore, these models quickly become outdated and require frequent updates. These models are also not suitable for the medium-term forecasting of real estate markets.
On the other hand, there are dynamic models [1, 35] taking into account the general state of the economy and designed to forecast and investigate the overall price situation in the real estate market. Nevertheless, these models are not intended for mass appraisal of real estate.
In order to overcome these shortcomings, the authors of the article offered to your attention have recently published works [36, 37], in which attempts were made to develop methods for creating complex models that have the properties of both static and dynamic models. These new models take into account both construction and operational characteristics of real estate objects as well as some parameters characterizing the changing economic situation in some region, country, and the world. Due to this, such models have become self-adaptable to time; i.e., they have learned to maintain their predictive capabilities regardless of the changing economic situation over time.
The aim of this paper is to further expand and develop the results of the previous studies [36, 37]. Our goal is to create a model that can be self-adaptive not only to time but also to space.
2. Materials and Methods
When creating a model for mass assessment and scenario forecasting of residential real estate markets in Russian cities, geographical, construction, operational, time, and macroeconomic factors were taken into account as input parameters.
The model included the following geographical factors: the city index (1: Moscow; 2: Saint Petersburg; 3: Yekaterinburg; 4: Perm; etc.), the geographical coordinates of a specific apartment house (latitude, longitude) identified using the Yandex service at an address specified, and the level of prestige of the house’s location on the geographical map of the city.
In this set of parameters, the city index, which links the estimated apartment to a specific city, is fundamentally new. The parameter that characterizes the degree of prestige of the house location on the geographical city map is also new. Let us look at this parameter in more detail, since this paper introduces it for the first time.
In order to take into account the house location prestige, professional appraisers often use the distance from a specific house to the city center. Sometimes, parameters that characterize transport accessibility, proximity to metro stations, parking lots, city squares, business and cultural centers, industrial enterprises, public toilets, etc. are considered as well. However, such parameters are subjective. For example, there are cities without some center. There are cities with several centers. Parking lots, squares, and cultural and business centers can vary in terms of their convenience and efficiency.
In this regard, we suggest using the so-called heat maps to assess the location of real estate objects. These heat maps are constructed as follows. In each city, many properties of a similar type are selected, for example, many two-room apartments of approximately the same size sold over a certain period of time. The coordinates of apartments are put on the map, and their market value is shown on the map in different colors. The zones where the most expensive apartments are located are shown in red gradually changing to colder colors as the cost of the apartment decreases. An example of a heat map of Yekaterinburg constructed in this way is shown in Figure 1.
... This result can be viewed as a difference in the spatial dependence and spatial heterogeneity between these medium-cost cities and highcost cities (Anselin 2013;Basu and Thibodeau 1998;Bitter et al. 2007). Finally, regarding the best machine learning methods, many studies in the literature have already demonstrated similar results with ensemble learning algorithms as their best predictors (e.g., McCluskey et al. 2014;Čeh et al. 2018;Mullainathan and Spiess 2017;Kok et al. 2017;Mayer et al. 2018;Baldominos et al. 2018) or with artificial neural networks outperforming the other methods (McCluskey et al. 2013;Yacim and Boshoff 2018;Abidoye et al. 2019). However, some other studies in the literature contrast the results with those of k-nearest neighbors (e.g., Isakson 1988;Borde et al. 2017) or support vector regression as the best predictors (e.g., Lam et al. 2009;Kontrimas and Verikas 2011;Huang 2019). ...
This paper reviews real estate price estimation in France, a market that has received little attention. We compare seven popular machine learning techniques by proposing a different approach that quantifies the relevance of location features in real estate price estimation with high and fine levels of granularity. We take advantage of a newly available open dataset provided by the French government that contains 5 years of historical data of real estate transactions. At a high level of granularity, we obtain important differences regarding the models’ prediction powers between cities with medium and high standards of living (precision differences beyond 70% in some cases). At a low level of granularity, we use geocoding to add precise geographical location features to the machine learning algorithm inputs. We obtain important improvements regarding the models’ forecasting powers relative to models trained without these features (improvements beyond 50% for some forecasting error measures). Our results also reveal that neural networks and random forest techniques particularly outperform other methods when geocoding features are not accounted for, while random forest, adaboost and gradient boosting perform well when geocoding features are considered. For identifying opportunities in the real estate market through real estate price prediction, our results can be of particular interest. They can also serve as a basis for price assessment in revenue management for durable and non-replenishable products such as real estate.
Purpose
Housing provision is a fundamental driver of national development and economic prosperity. However, housing affordability remains a critical problem worldwide, affecting countries regardless of their level of development, whether developed, developing or underdeveloped. The housing price index (HPI) measures trends in residential property prices and provides valuable insights into market dynamics. This study aims to model and forecast the HPI in Malaysia. Accurate HPI predictions are essential for informing policy decisions, guiding investment strategies and ensuring sustainable housing market development.
Design/methodology/approach
The autoregressive integrated moving average (ARIMA) model was used to analyse historical housing price data for modelling and forecasting housing prices for different types. The accuracy and reliability of the models and forecasts were assessed using key statistical metrics, including variance (s²), log-likelihood, Akaike information criterion, corrected Akaike information criterion and Bayesian information criterion.
Findings
The findings demonstrate that the ARIMA model improves accuracy in capturing seasonal and trend components in the Malaysian housing market, offering actionable insights for strategic decision-making. In addition, this research presents point and annual changes across different housing types that seek to offer a detailed view of price movements. By providing a reliable five-year forecast of the HPI, this study enables more informed decision-making in housing policy, investment strategy and financial planning.
Practical implications
Policymakers can use these insights to stabilise housing markets, whereas investors and developers can anticipate market movements for better capital allocation. Homeowners and prospective buyers will benefit from understanding future price trends, allowing for more strategic decisions regarding property purchases or sales.
Originality/value
The originality of the study lies in its ability to empirically forecast housing prices for various types of housing for the next five years. This provides valuable guidance for housing-related decisions. This research contributes to the growing literature on real estate forecasting in emerging markets and underscores the relevance of advanced time series models in navigating the complexities of dynamic housing markets.
The Chinese real estate market has grown at such a quick rate over the last few decades, up to the current falling patterns that began at the end of 2021. This difficulty has made it more difficult for the government and investors to predict future property prices effectively. This has occurred as a result of the state of the economy at the moment. In this research, we examine the monthly residential property prices in Hangzhou City, Zhejiang Province, China, using Gaussian process regressions with a variety of kernels and basis functions. This research spans the months of January 2009 through July 2024. We use estimated models in our forecasting efforts. A combination of cross-validation and Bayesian optimisations is used to train these models. The prices that would be seen outside of the sample from June 2021 to July 2024 were successfully predicted by the generated models. These models have an accuracy of 1.0419 per cent for the relative root mean square error. It is plausible that our findings may be used alone or in combination with further projections to formulate theories about fluctuations in residential real estate prices and to conduct supplementary policy analysis.
This paper aims at developing forecasting models to predict the sales of a real estate project in India using variables like amenities, neighbourhood charac-teristics, price and registration of real estate project. The predictive models include multiple linear regression (MLR)/ordinary least squares (OLS), random forest (RF) and artificial neural network (ANN) to discuss the efficacy and popularity of real estate forecasting models, their results and biases. Discussed models are divided into four categories: macro-economic forecasting models, micro-economic forecasting models, time-series models and spatial hedonic models. Hedonic regression models are one of the most common models adopted by researches followed by artificial neural network (ANN) and time-series models. Random forest and ANN models suggest that nonlinear algorithms show better performance in the prediction. Use of ANN outperforms the use of other methods such as the OLS. Prediction accuracy is highest in neural network, followed by random forest and lastly by OLS for the model tested. Finally, the inference from this study is that the forecasting models require recalibration over time and require variables updating for forecasting effi-cacy. Limitation of their use and benchmarking depends on country, location, culture and context, with limited portability.
Keywords Artificial neural network (ANN) ·Hedonic models ·Macro-economic models ·Micro-economic models ·Real estate forecasts ·Multiple linear regression (MLR) ·Ordinary least square (OLS) ·Random forest
INTRODUCTION:-In the realm of building management, safety and security stand as pillars of paramount importance. From commercial skyscrapers to residential complexes, ensuring the well-being of occupants and safeguarding valuable assets is a top priority. Fortunately, advancements in technology have ushered in a new era of safety and security measures, enabling buildings to be smarter, safer, and more secure than ever before. In this article, we explore how these advancements are reshaping the landscape of building management and fortifying our foundations.
Supervisory Technology (SupTech) combines artificial intelligence (AI) with machine learning (ML) for the specific purpose of supervision of the financial markets. SupTech offers transformative solutions for the regulation, monitoring and surveillance of financial markets. This article presents a literature review examining the role of SupTech in enhancing market surveillance. Using the PSALSAR framework, we conduct a Systematic Literature Review (SLR) of 55 articles published between 1999 and 2022. The review highlights the potential of SupTech to navigate the complexities of the financial landscape, and its versatile applications for financial supervisors and regulators, stock exchanges, intermediaries, and investors. Our findings indicate that SupTech models can effectively detect instances of market manipulations, insider trading, and fraud. These models also showcase exceptional potential to forecast crises in markets, economies, corporations, and in the banking sector. They are useful in empowering policymakers and regulators to make informed decisions. The study provides a strong foundation for future research and policy development to explore and deploy the potential of SupTech, paving the way for greater transparency, efficiency, and resilience in financial markets.
Purpose
The Chinese housing market has gone through rapid growth during the past decade, and house price forecasting has evolved to be a significant issue that draws enormous attention from investors, policy makers and researchers. This study investigates neural networks for composite property price index forecasting from ten major Chinese cities for the period of July 2005–April 2021.
Design/methodology/approach
The goal is to build simple and accurate neural network models that contribute to pure technical forecasts of composite property prices. To facilitate the analysis, the authors consider different model settings across algorithms, delays, hidden neurons and data spitting ratios.
Findings
The authors arrive at a pretty simple neural network with six delays and three hidden neurons, which generates rather stable performance of average relative root mean square errors across the ten cities below 1% for the training, validation and testing phases.
Originality/value
Results here could be utilized on a standalone basis or combined with fundamental forecasts to help form perspectives of composite property price trends and conduct policy analysis.
This study investigates the relationship between the weighted average interest rates applied to housing loans, housing construction labor cost and housing unit prices, and the housing price index for new dwellings for the period 2015-2021. In the time series, a structural break was detected in February 2019. In the results of the regression analysis, differences were determined for the January 2015-January 2019 period and the February 2019-August 2021 period.
The real estate market in China has been growing rapidly during the past decade, with different property price patterns across various regions. Among different types of properties, prices of retail properties have not been sufficiently analyzed. This study focuses on forecasting problems of price indices of retail properties across ten major cities in China from 2005 to 2021 through monthly data and Gaussian process regression models. We examine ten kernels, four basis functions, and two predictor standardization options for constructing the forecast model through Bayesian optimization and cross-validation for each price index. The ten models built lead to accurate out-of-sample forecast results for a two-year window from 2019 to 2021. Relative root mean square errors across the ten indices range from 0.0113% to 0.4835%. Market participants and policymakers could utilize forecast models here for assessing retail property prices.
Testing a model in property evaluation can be a difficult task due to the large variety of these models. The most popular models used in valuation are regression and neural networks. This paper applied a systematic review study and presents 11 types of regression models and 9 types of neural network models applied in real estate valuation. Our aim is to provide a tool for model selection applied in real estate valuation. The selection criteria were based on their applicability, user preferences and price estimation performance. The findings were slightly different from our expectations. Multi-Layer Perceptron (MLP) and Multiple Linear Regression (GLM) are the most applied and popular models in valuation.
Rapid growth has been seen in the Chinese housing market during the past decade, and housing price forecasts have become more significant to investors and policy makers. In this study, we explore neural networks for retail property price index forecasts from ten major Chinese cities for July 2005–April 2021. Our goal is to build simple and accurate neural networks to contribute to purely technical forecasts of the Chinese retail property market. To facilitate the analysis, we examine different model settings based on algorithms, delays, hidden neurons, and data splitting ratios and construct a simple neural network with five delays and two hidden neurons. This leads to a rather stable performance of about 1.6% average relative root mean square error across the ten cities for the training, validation, and testing phases. Results here can be used on a stand-alone basis or combined with fundamental forecasts to form perspectives of retail property price trends and conduct policy analyses.
Urban regeneration interventions and strategies for the restoration of eco-systemic services, in addition to generating a range of community benefits, lead to an increase in property values in the neighbourhood. This study aims to investigate how redevelopment affects residential house prices.The aim of the paper is twofold. First the predictive potential of hedonic models is compared with that of Artificial Intelligence (AI) and Machine Learning (ML) approaches. Then, we characterise a novel Artificial Neural Network (ANN) model, which is still scarcely used to forecast how housing prices vary due to changes in the quality of the urban environment. The model includes among the variable inputs a series of environmental factors rarely considered in AI-ML approaches. The defined ANN is intended to support traditional approaches in order to provide planners and decision-makers with a more complete set of information on real estate trends.Applications to real case studies will allow to validate the model, as well as to identify the environmental variables that most significantly influence residential property values.KeywordsUrban renewalHousing pricesArtificial Neural NetworksMultiple Regression Analysis
This study attempts to assess macroeconomic considerations in the innovation and adoption of self-
help or incremental affordable housing finance across the sub-Saharan African markets. This is with the view of
improving decision-making for financiers, product developers, and lenders. The desktop survey was used to gather
relevant information relative to the positions in research on the role of macroeconomic indicators in the growth
of investment in informal financing solutions. This descriptive content analysis covers literature spanning a period
of 1995-2018 (23 years). The studies reviewed suggest that macroeconomic influences can explain the market’s
growth with regards to financing incremental housing. This study adds to the discuss by exploring the
macroeconomic influences that may be responsible for the growth or decline of investment of financing for
incremental or informal housing in Africa. This provides policymakers and lenders with insight into the trends
that drive their decisions. This could potentially improve how investment decisions are evaluated and made.
#housingfinance #incrementalhousing #Africanhousing #Affordablehousing #HousingSolutions
Purpose
India is one of those countries that are severely affected by the COVID-19 pandemic. With the upsurge in the cases, the country recorded high unemployment rates, economic uncertainties and slugging growth rates. This adversely affected the real estate sector in India. As the relation of the housing market with the gross domestic product is quite lasting thus, the decline in housing prices has severely impacted the economic growth of the nation. Hence, the purpose of this paper is to gauge the asymmetric impact of COVID-19 shocks on housing prices in India.
Design/methodology/approach
Studies revealed the symmetric impact of macroeconomic variables, and contingencies on housing prices dominate the literature. However, the assumption of linearity fails to apprehend the asymmetric dynamics of the housing sector. Thus, the author uses a nonlinear autoregressive distributed lag model to address this limitation and test the existence of short- and long-run asymmetry.
Findings
The findings revealed the long- and short-run asymmetric impact of the COVID-19 outbreak and the peak of the COVID-19 on housing prices. The results indicate that the peak of COVID-19 had a greater impact on housing prices in comparison to the outbreak of COVID-19. This can be explained as prices will revert to normal at a speed of 0.978% with the decline in the number of COVID-19 cases. Whereas the housing prices rise at a rate of 0.714 as a result of government intervention to deal with the ill effects of the COVID-19 outbreak. Moreover, it can be inferred that both the outbreak and peak of COVID-19 will lead to a minimal decline in housing prices, while with the decline in the number of cases and reduction in the impact of the outbreak of COVID, the housing prices will rise at an increasing rate.
Originality/value
To the best of the authors’ knowledge, this is the first study to understand the impact of the outbreak and peak of COVID-19 on the housing prices separately.
For years, the hedonic regression model has dominated housing price research worldwide. However, the hedonic regression model suffered from the problem of over-simplification and heterogeneity. Machine learning has become a hot method in housing price prediction in recent years. The machine learning method in predicting housing prices is more accurate and precise than the traditional methods. This paper introduced three regression methods in housing price prediction: the traditional hedonic regression model, Google AutoML and Microsoft AutoML. It reviewed the factors that affected housing prices in literature and used the dataset of the housing price in Beijing in Kaggle to study the factors affected the housing price in Beijing. The results showed that Google AutoML had the best performance in predicting housing prices in Beijing. It had the highest R square (0.820) and the least RMSE and MAE. The average housing price in a community was the most important feature that impacted housing price prediction. Number of days open for sale and geographical location ranked the second and the third most important features in predicting the housing price.KeywordsMachine learningBeijingAutoML
Purpose
The assessment of the Real Estate (RE) prices depends on multiple factors that traditional evaluation methods often struggle to fully understand. Housing prices, in particular, are the foundations for a better knowledge of the Built Environment and its characteristics. Recently, Machine Learning (ML) techniques, which are a subset of Artificial Intelligence, are gaining momentum in solving complex, non-linear problems like house price forecasting. Hence, this study deployed three popular ML techniques to predict dwelling prices in two cities in Italy.
Design/methodology/approach
An extensive dataset about house prices is collected through API protocol in two cities in North Italy, namely Brescia and Varese. This data is used to train and test three most popular ML models, i.e. ElasticNet, XGBoost and Artificial Neural Network, in order to predict house prices with six different features.
Findings
The models' performance was evaluated using the Mean Absolute Error (MAE) score. The results showed that the artificial neural network performed better than the others in predicting house prices, with a MAE 5% lower than the second-best model (which was the XGBoost).
Research limitations/implications
All the models had an accuracy drop in forecasting the most expensive cases, probably due to a lack of data.
Practical implications
The accessibility and easiness of the proposed model will allow future users to predict house prices with different datasets. Alternatively, further research may implement a different model using neural networks, knowing that they work better for this kind of task.
Originality/value
To date, this is the first comparison of the three most popular ML models that are usually employed when predicting house prices.
A relatively high level of precision is required in real estate valuation for investment purposes. Such estimates of value which is carried out by real estate professionals are relied upon by the end-users of such financial information having paid a certain fee for consultation hence leaving little room for errors. However, valuation reports are often criticised for their inability to be replicated by two or more valuers. Hence, stirring to a keen interest within the academic cycle leading to the need for exploring avenues to improve the price prediction ability of the professional valuer. This study, therefore, focuses on overcoming these challenges by introducing an integrated approach that combines ANFIS with ANN termed ANFIS-AN, thereby having a reiteration in terms of ANN application to fortify price predictability. Using 255 property data alongside 12 variables, the ANFIS-AN model was adopted and its outcome was compared with that of ANN. Finally, the results were subjected to 3 different error testing models using the same training and learning benchmarks. The proposed model’s RMSE gave 1.413169, while that of ANN showed 9.942206. Similarly, using MAPE, ANN recorded 0.256438 while ANFIS-AN had 0.208324. Hence, ANFIS-AN’s performance is laudable, thus a better tool over ANN.
Purpose-By considering the rapid and continuous increase of housing prices in Turkey recently, this study aims to examine the determinants of the residential property price index (RPPI). In this context, a total of 12 explanatory (3 macroeconomic, 8 markets and 1 pandemic) variables are included in the analysis. Moreover, the residential property price index for new dwellings (NRPPI) and the residential property price index for old dwellings (ORPPI) are considered for robustness checks.
Design/methodology/approach-A quantile regression (QR) model is used to examine the main determinants of RPPI in Turkey. A monthly time series data set for the period between January 2010 and October 2020 is included. Moreover, NRPPI and ORPPI are examined for robustness.
Findings-Predictions for RPPI, NRPPI and ORPPI are carried out separately at the country (Turkey) level. The results show that market variables are more important than macroeconomic variables; the pandemic and rent have the highest effect on the indices; The effects of the explanatory variables on housing prices do not change much from low to high levels, the COVID-19 pandemic and weighted average cost of funding have a decreasing effect on indices while other variables have an increasing effect in low quantiles; the pandemic and monetary policy indicators have a negative and significant effect in low quantiles whereas they are not effective in high quantiles; the results for RPPI, NRPPI and ORPPI are consistent and robust.
The Indian textile industry is one of the notable contributors to the national economy. In this modern and connected world, timely, accurate, and informed decisions are the key to sustainable growth. In the present research work, we proposed a Multilayer Perceptron (MLP) approach, developed efficient forecasting models using it for the Wholesale Price Index (WPI) of all the twenty-five individual items of the manufacture of the textiles group of India. Upon comparison, the proposed MLP approach with the ARIMA and Holt-Winters approaches – two well-known time-series forecast approach, we observed all the three demonstrate reliable and acceptable model accuracy (forecast accuracy) based on Theil’s U statistics. We also observed the proposed MLP exhibited high model (forecast) accuracy with MAPE ≤ 10 for 96% of the indices. Further, the proposed MLP outshines others in terms of the maximum number of counts of the lowest MAPE, RMSE, and Theil’s U statistics. Therefore, the proposed MLP approach can be an alternative approach for forecasting these wholesale price indices.
Purpose
The purpose of this study is to compare the predictive performance of the hedonic multivariate regression model with the probabilistic neural network (PNN)-based general regression neural network (GRNN) model of housing prices in “Pune-India.”
Design/methodology/approach
Data on 211 properties across “Pune city-India” is collected. The price per square feet is considered as a dependent variable whereas distances from important landmarks such as railway station, fort, university, airport, hospital, temple, parks, solid waste site and stadium are considered as independent variables along with a dummy for amenities. The data is analyzed using a hedonic type multivariate regression model and GRNN. The GRNN divides the entire data set into two sets, namely, training set and testing set and establishes a functional relationship between the dependent and target variables based on the probability density function of the training data (Alomair and Garrouch, 2016).
Findings
While comparing the performance of the hedonic multivariate regression model and PNN-based GRNN, the study finds that the output variable (i.e. price) has been accurately predicted by the GRNN model. All the 42 observations of the testing set are correctly classified giving an accuracy rate of 100%. According to Cortez (2015), a value close to 100% indicates that the model can correctly classify the test data set. Further, the root mean square error (RMSE) value for the final testing for the GRNN model is 0.089 compared to 0.146 for the hedonic multivariate regression model. A lesser value of RMSE indicates that the model contains smaller errors and is a better fit. Therefore, it is concluded that GRNN is a better model to predict the housing price functions. The distance from the solid waste site has the highest degree of variable senstivity impact on the housing prices (22.59%) followed by distance from university (17.78%) and fort (17.73%).
Research limitations/implications
The study being a “case” is restricted to a particular geographic location hence, the findings of the study cannot be generalized. Further, as the objective of the study is restricted to just to compare the predictive performance of two models, it is felt appropriate to restrict the scope of work by focusing only on “location specific hedonic factors,” as determinants of housing prices.
Practical implications
The study opens up a new dimension for scholars working in the field of housing prices/valuation. Authors do not rule out the use of traditional statistical techniques such as ordinary least square regression but strongly recommend that it is high time scholars use advanced statistical methods to develop the domain. The application of GRNN, artificial intelligence or other techniques such as auto regressive integrated moving average and vector auto regression modeling helps analyze the data in a much more sophisticated manner and help come up with more robust and conclusive evidence.
Originality/value
To the best of the author’s knowledge, it is the first case study that compares the predictive performance of the hedonic multivariate regression model with the PNN-based GRNN model for housing prices in India.
Accurate price evaluation of real estate is beneficial for many parties involved in real estate business such as real estate companies, property owners, investors, banks, and financial institutes. Artificial Neural Networks (ANNs) have shown promising results in real estate price evaluation. However, the performance of ANNs greatly depends upon the settings of their hyperparameters. In this paper, we apply and optimize an ANN model for real estate price prediction in Helsinki, Finland. Optimization of the model is performed by fine-tuning hyper-parameters (such as activation functions, optimization algorithms, etc.) of the ANN architecture for higher accuracy using the Bayesian optimization algorithm. The results are evaluated using a variety of metrics (RMSE, MAE, R2) as well as illustrated graphically. The empirical analysis of the results shows that model optimization improved the performance on all metrics (reaching the relative mean error of 8.3%).
Purpose
This paper aims to investigate asymmetric pricing behaviour and impact of coronavirus (Covid-19) pandemic shocks on house price index (HPI) of Turkey and Kazakhstan.
Design/methodology/approach
Monthly HPIs and consumer price index (CPI) data ranges from 2010M1 to 2020M5 are used. This study uses a nonlinear autoregressive distributed lag model for empirical analysis.
Findings
The findings of this study reveal that the Covid-19 pandemic exerted both long-run and short-run asymmetric relationship on HPI of Turkey while in Kazakhstan, the long-run impact of Covid-19 pandemic shock is symmetrical long-run positive effect is similar in both HPI markets.
Research limitations/implications
The main limitations of this study are the study scope and data set due to data constraint. Several other macroeconomic variables may affect housing prices; however, variables used in this study satisfy the focus of this study in the presence of data constraint. HPI and CPI variables were made available on monthly basis for a considerably longer period which guaranteed the ranges of data set used in this study.
Practical implications
Despite the limitation, this study provides necessary information for authorities and prospective investors in HPI to make a sound investment decision.
Originality/value
This is the first study that rigorously and simultaneously examines the pricing behaviour of Turkey and Kazakhstan HPIs in relation to the Covid-19 pandemic shocks at the regional level. HPI of Kazakhstan is recognized in the global real estate transparency index but the study is rare. The study contributes to regional studies on housing price by bridging this gap in the real estate literature.
The appraisal of large amounts of properties is often entrusted to Automated Valuation Models (AVM). At one time, only econometric models were used for this purpose. More recently, also machine learning models are used in mass appraisal techniques. The literature has devoted much attention to assessing the performance capabilities of these models. Verification tests first train a model on a training set, then measure the prediction error of the model on a set of data not met before: the testing set. The prediction error is measured with an accuracy indicator. However, verification on the testing set alone may be insufficient to describe the model’s performance. In addition, it may not detect the existence of model bias such as overfitting. This research proposes the use of cross validation to provide a more complete and effective evaluation of models. Ten-fold cross validation is used within 5 models (linear regression, regression tree, random forest, nearest neighbors, multilayer perception) in the assessment of 1,400 properties in the city of Turin. The results obtained during validation provide additional information for the evaluation of the models. This information cannot be provided by the accuracy measurement when considered alone.
Purpose
The paper aims to investigate the application of particle swarm optimisation and back propagation in weights optimisation and training of artificial neural networks within the mass appraisal industry and to compare the performance with standalone back propagation, genetic algorithm with back propagation and regression models.
Design/methodology/approach
The study utilised linear regression modelling before the semi-log and log-log models with a sample of 3,242 single-family dwellings. This was followed by the hybrid systems in the selection of optimal attribute weights and training of the artificial neural networks. Also, the standalone back propagation algorithm was used for the network training, and finally, the performance of each model was evaluated using accuracy test statistics.
Findings
The study found that combining particle swarm optimisation with back propagation in global and local search for attribute weights enhances the predictive accuracy of artificial neural networks. This also enhances transparency of the process, because it shows relative importance of attributes.
Research limitations/implications
A robust assessment of the models’ predictive accuracy was inhibited by fewer accuracy test statistics found in the software. The research demonstrates the efficacy of combining two models in the assessment of property values.
Originality/value
This work demonstrated the practicability of combining particle swarm optimisation with back propagation algorithms in finding optimal weights and training of the artificial neural networks within the mass appraisal environment.
Owning residential real estate is very important not only for satisfying the need for shelter—a fundamental human right—but also it is considered as a good investment. It is true that real estates are heterogeneous because of their various structural, positional characteristics, as well as their local conditions, all of which affect the value of real estates. These are among the reasons for undertaking studies to determine the degree of importance and the parameters affecting the value of residential real estate. By using one of the widely used methods, hedonic pricing method (HPM), this study seeks to determine the structural and environmental characteristics that are effective on residential real estate sale prices in Artvin city center. With the highest number of residential real estate sales in 2015, the Orta neighborhood, in the Central District of Artvin, was chosen as the study area. The relationship between 18 parameters were analyzed, on the basis of the actual sales value, acquired from the purchasers of 81 residential properties in this neighborhood and taking into account the structure of Artvin city. In this context, using the functional form of HPM, semi-logarithmic model was generated. When examining the model, it was determined that four out of the supposedly effective 18 parameters were statistically significant and they were able to explain 84% of the variations on price. Moreover, the parameters of floor area, age, distance to primary school, and distance to city center were effective on the sales prices in the semi-logarithmic model. In addition, the results revealed that the structural characteristics are more influential on real estate prices than the environmental and accessibility features within Orta neighborhood in Artvin.
High levels of car ownership have major impacts on congestion and thus the mobility, accessibility, heath and liveability in cities. Increasing car ownership is already reflected in high congestion levels in South African cities and does not appear to be reducing, despite policy interventions. The factors that drive the high car ownership intentions thus need to be investigated, so that policy efforts can be appropriately directed. The study aimed at investigating the car ownership intentions of students, as being most likely to drive car sales in the future, with the purpose of understanding the factors underlying the high desire to own a car. The study finds that although costs are the main barrier to market entry, and that most students intend to purchase a car as soon as they can afford it. These intentions are largely driven by the view that the quality of public transport constrains the movement of people and does not provide a travel alternative that is considered to be a reasonable alternative to the car, as indicated by the view that cars are a necessity. The study finds that although there are differences in the valuation of public and alternative modes of transport, based on demographic elements, familiarity with car usage and psychosocial factors, most students intend to own a car as the best means of travel, with little seeming to moderate the decision. The poor valuation of public and alternative transport suggests however that, whilst other measures to curb car use and promote public transport may have value, only significant service level improvements in public transport is likely to drive real behaviour change.
The real estate market in Hong Kong plays an important role in its economy. The property prices have been increasing a lot since 2009, which have become a major concern. However, few studies have been done to forecast the property price indices in Hong Kong. In this paper, two grey models, GM(1,1) and GM(0,N), are introduced for the forecasting. The results show that GM(1,1) has a better performance when forecasting with stable trend data, while GM(0,N) is more suitable for forecasting data in fluctuating trend. The sensitivity analysis for GM(0,N) shows that Population(POP) and Best Lending Rate(BLR) are significantly sensitive factors for data in stable trend. While for the fluctuating data, sensitivity of each factor presents uncertainties. This study also compares the forecasting performance of grey models with the ANN model and ARIMA model. The study demonstrates that grey models are more suitable for forecasting the Hong Kong property price indices than others.
Machine learning techniques are applied to the analysis of real data on the new housing market of Santiago, Chile. The objective is to compare the predictive performance of the Neural Network, Random Forest and Support Vector Machine approaches with traditional Ordinary Least Squares Regression. The database for our analysis consists of a sample of 16,472 price records for new housing units or residential properties within the area covered. The results of the analysis show that Random Forest performed better than the other models in modeling housing prices. More generally, we conclude that machine learning techniques can provide a useful set of tools for acquiring information on housing markets.
The Artificial Neural Network (ANN) technique has been applied and found useful for solving forecasting problems in different property markets around the world. However, the trend of ANN's application in property valuation research has been undocumented. Therefore, this study aims to systematically review the extant literature on the application of ANN to property valuation. The analysis of the retrieved articles revealed that a seminal study in this area was reported in 1991. To date, the technique has continued to gain popularity amongst real estate researchers. Most of the reviewed articles originated from developed countries, particularly the US and the UK, although a few studies emanated from emerging economies. Most of the authors that contributed to the publications are affiliated to university faculties and most of the studies found ANN to have outperformed other appraisal techniques, in terms of accuracy. The gaps identified in this study need to be addressed in order to achieve sustained growth in property appraisal practice on a global scale.
Purpose
This study aims to analyze whether urban tourism affects Berlin housing rents. Urban tourism is of considerable economic importance for many urban destinations and has developed very strongly over the past few years. The prevailing view is that urban tourism triggers side-effects, which affect the urban housing markets through a lack of supply and increasing rents. Berlin represents Germany’s largest rental market and is particularly affected by growing urban tourism and increasing rents.
Design/methodology/approach
The paper considers whether urban tourism hotspots affect Berlin’s housing rents, using two hedonic regression approaches, namely, conventional ordinary least squares (OLS) and generalized additive models (GAM). The regression models incorporate housing characteristics as well as several distance-based measures. The research considers tourist attractions, restaurants, hotels and holiday flats as constituents of tourism hotspots and is based on a spatial analysis using geographic information systems (GIS).
Findings
The results can be regarded as a preliminary indication that rents are, indeed, affected by urban tourism. Rents seem to be positively correlated with the touristic attractiveness of a particular location, even if it is very difficult to accurately measure the real quantity of the respective effects of the urban tourism amenities, as the various models show. GAM outperforms the results of OLS and seems to be more appropriate for spatial analysis of rents across a city.
Originality/value
To the best of the authors’ knowledge, the paper provides the first empirical analysis of the effects of urban tourism hotspots on the Berlin housing market.
The direction and mechanisms of the relationship between airport traffic volumes and property prices are somewhat unclear in the literature. This study adds to that body of knowledge by empirically investigating the role of airports as an essential driver of economic activity by creating employment and facilitating air travel between destinations. The two-stage least-squares (2SLS) approach is employed to investigate the link between house prices and the airport traffic volumes of New Zealand’s three key regions and airports (Auckland, Canterbury/Christchurch and Wellington) from July 2004 to December 2014. The empirical findings of the study suggest that airport traffic volumes positively and significantly influence the urban house prices of New Zealand’s three major regions.
This paper attempts to build an artificial neural network that can estimate the median price of a home in a neighborhood described by forty demographic attributes in areas of Dhaka City namely Dhanmondi, Baridhara, Gulshan, Mirpur, Uttara and Old Dhaka. A structured questionnaire was used to collect the relevant data and the housing data sets was used to develop constant quality price indices using traditional econometric techniques and using neural networks incorporating genetic algorithms. Factors including house size, house age, house type, number of bedrooms, number of bathrooms, number of garages, amenities around the house and geographical location are considered. The analysis indicates that neural networks act as real alternative to the econometric methods. In this paper thirteen conditional attributes have been considered to estimate the house rent. This paper provides some indicative policy guidelines to handle the house rent problem in the Dhaka City and suggests that a rent controller should be appointed for each ward and maximum rents for particular areas should be gazette and the rents should be paid through banks.
This paper builds a house prices forecasting model for private residential houses in HongKong, based on general macroeconomic indicators, housing related data and demographicfactors for the period of 1980 to 2001. A reduce form economic model has been derivedfrom a multiple regression analysis where three sets and eight models were derived foranalysis and comparison. It is found that household income, land supply, population andmovements in the Hang Seng Index play an important role in explaining house pricemovements in Hong Kong. In addition, political events, as identified, cannot be ignored.However, the results of the models are unstable. It is suggested that the OLS may nota best method for house prices model in Hong Kong situation. Alternative methods aresuggested.
An accurate measurement of the impacts of external shocks on construction demand will enable construction industry policymakers and developers to make allowances for future occurrences and advance the construction industry in a sustainable manner. This paper aims to measurethe dynamic effects of the late 2000s global financial crisis on the level of demand in the Australian construction industry. The vector error correction (VEC) model with intervention indicators is employed to estimate the external impact from the crisis on a macro-level construction economic indicator, namely construction demand. The methodology comprises six main stages to produce appropriate VEC models that describe the characteristics of the underlying process. Research findings suggestthat overall residential and non-residential construction demand were affected significantly by the recent crisis and seasonality. Non-residentialconstruction demand was disrupted more than residential construction demand at the crisis onset. The residential constructionindustry is more reactive and is able to recover faster following the crisis in comparison with the non-residential industry. The VEC model with intervention indicators developed in this study can be used as an experiment for an advanced econometric method. This can be used to analyse the effects of special eventsand factors not only on construction but also on other industries.
This paper estimates the real estate cycle in Thailand. From the estimated results, we find that duration of the expansion period in the real estate cycle in Thailand was approximately 25.25 months while the contraction period lasted much longer (44.00 months). The duration of the trough-to-trough cycle is estimated to be approximately 69.25 months. The significant leading indicators for the real estate cycle are construction price index, money supply (M2), property stock index and post-credit finance. Compared to Thailand's economic cycle, the real estate cycle leads the trough and the peak in the business/economic cycle by approximately 14.3 months and 20.3 months respectively. In expansion periods, the real estate cycle is always found to lead the business/economic cycle. However, it is not clear that in contraction periods the real estate cycle leads the business cycle. This finding differs from that of previous studies. We found that real estate crises led to economic crises in the early 1980s and in 1997, while in other contraction periods it was an economic recession that led to a contraction in the real estate sector.
Spatial autocorrelation is commonly found in the Hedonic Pricing model for real estate prices, but little attention has been paid to studying its time-varying characteristic. Most previous studies consider that spatial autocorrelation is a time-invariant parameter with a pre-determined time window of previous nearby transactions. The primary objective of this study is to examine how spatial autocorrelation varies with the length of this time window. Since nearby transactions for a subject property may pose different extents of spatial autocorrelation when different periods of past sales are referred to, the spatial autocorrelation will therefore vary with the reference period under consideration. We hypothesize that the reference period of past sales affects the geographical boundary within which buyers and sellers search for price information, and thus, spatial auto-correlation. The more recent sales should have a stronger impact on the transaction prices of the subject property than the old ones, and should therefore induce stronger spatial auto-correlation.
A Spatial Hedonic Pricing (SHP) model is proposed to test our hypothesis. The SHP model generalizes traditional spatial autoregressive models by making the spatial weight time-dependent. Based on 15,500 transactions of residential units in Taikooshing, Hong Kong from 1992 to 2006, we conclude that while positive spatial autocorrelation is present in housing prices, the spatial autocorrelation varies with time and its magnitude increases when the reference period of past sales becomes shorter. These are new findings that contribute to the literature on spatial econometrics. Our results not only reveal the spatially dependent price formation process in the real estate market, but also have practical applications on the hedonic modelling of real estate prices for mass valuation and index construction.
Barbados is one of the most popular destinations in the Caribbean. Historically most visitors were accommodated in hotels. In the past decade, however, the proportion of luxury rooms has declined in proportion to the expansion in the second home rental market. The behavior of rental prices for villas on the island, nevertheless, is not well understood. Using observations on 322 villas and cottages, this article estimates a hedonic regression of the prices for these properties. Prices are modeled as functions of structural, neighborhood, and environmental factors. The findings from the article suggest that the rental prices for villas and cottages on the island are largely because of unobserved neighborhood characteristics and structural features related to the property. The study contributes to the literature by providing an assessment of the villa market in a largely tourism-dependent economy and also proposes an approach for measuring unobserved neighborhood characteristics.
In this paper, a model based on Artificial Neural Network (ANN) has been applied to real estate appraisal. Moreover, an evaluation of ANN performances in estimating the sale price of residential properties has been carried out. Artificial Neural Networks (ANNs) are useful in modelling input-output relationships learning directly from observed data. This capability can be very useful in complex systems like the real estate ones where motivations, tastes and budget availability often do not follow rational behaviours. This study also analyses the impact of such key environmental conditions that represent a problem related to many industrial cities where pollution and landscaping consequences affect the real estate market and residential location choices. We have considered a set of asking price's houses collected in the urban area of Taranto (Italy) where the biggest European steel factory and the 2nd industrial harbour are located.
Purpose
– The purpose of this paper is to apply boosted regression trees (BRT) to a heterogeneous data set of residential property drawn from a jurisdiction in Malaysia, with the objective to evaluate its application within the mass appraisal environment in Malaysia. Machine learning (ML) techniques have been applied to real estate mass appraisal with varying degrees of success.
Design/methodology/approach
– To evaluate the performance of the BRT model two multiple regression analysis (MRA) models have been specified (linear and non-linear). One of the weaknesses of traditional regression is the need to a priori specify the functional form of the model and to ensure that all non-linearities have been accounted for. For a BRT model the algorithm does not require any predetermined model or variable transformations, making the process much simpler.
Findings
– The results show that the BRT model outperformed the MRA-specified models in terms of the coefficient of dispersion and mean absolute percentage error. While the results are encouraging, BRT models still lack transparency and suffer from the inability to translate variable importance into quantifiable variable effects.
Practical implications
– This paper presents a useful alternative modelling technique, BRT, for use within the mass appraisal environment in Malaysia. Its advantages include less intensive data cleansing, no requirement to specify the predictive underlying model, ability to utilise categorical variables without the need to transform them and not as data hungry, as for example, MRA.
Originality/value
– This paper adds to the knowledge in this area by applying a relatively new ML model, BRT to residential property data from a jurisdiction in Malaysia. BRT has shown promise as a strong predictive model when applied in other disciplines; therefore this research empirically tests this finding within real estate valuation.
Some property buyers consider green building features such as balconies as an important element in purchasing a residential property. We investigate if there is any effect of a balcony, as a green feature, on property price. We find that balconies exert a strong significant and positive effect on residential property price. Property buyers are willing to pay a significant sale price premium of 4% more for units with balconies. We also find that balconies with different views are valued differently by buyers: a larger premium of 6.9% on a unit with a sea view balcony and 1.5% on a unit with mountain view. However, units with building views are not appreciated by buyers. It implies that properly architecturally designed balconies would add more value to properties and enhance their competitiveness.
The paper investigates the characteristics of house price
dynamics and the role of institutional factors in nine Asia-
Pacific economies during 1993–2006. On average, house prices
tend to be more volatile in markets with lower supply elasticity
and a more flexible business environment. At the national
level, the current run-up in house prices mainly reflects adjustment
to improved fundamentals rather than speculative housing
bubbles. However, evidence of bubbles does exist in some
market segments.
Inaccuracies in property valuation is a global problem. This could be attributed to the adoption of valuation approaches, with the hedonic pricing model (HPM) being an example, that are inaccurate and unreliable. As evidenced in the literature, the HPM approach has gained wide acceptance among real estate researchers, despite its shortcomings. Therefore, the present study set out to evaluate the predictive accuracy of HPM in comparison with the artificial neural network (ANN) technique in property valuation. Residential property transaction data were collected from registered real estate firms domiciled in the Lagos metropolis, Nigeria, and were fitted into the ANN model and HPM. The results showed that the ANN technique outperformed the HPM approach, in terms of accuracy in predicting property values with mean absolute percentage error (MAPE) values of 15.94 and 38.23%, respectively. The findings demonstrate the efficacy of the ANN technique in property valuation, and if all the preconditions of property value modeling are met, the ANN technique is a reliable valuation approach that could be used by both real estate researchers and professionals.
This research essay highlights the need to integrate predictive analytics into information systems research and shows several concrete ways in which this goal can be accomplished. Predictive analytics include empirical methods (statistical and other) that generate data predictions as well as methods for assessing predictive power. Predictive analytics not only assist in creating practically useful models, they also play an important role alongside explanatory modeling in theory building and theory testing. We describe six roles for predictive analytics: new theory generation, measurement development, comparison of competing theories, improvement of existing models, relevance assessment, and assessment of the predictability of empirical phenomena. Despite the importance of predictive analytics, we find that they are rare in the empirical IS literature. Extant IS literature relies nearly exclusively on explanatory statistical modeling, where statistical inference is used to test and evaluate the explanatory power of underlying causal models, and predictive power is assumed to follow automatically from the explanatory model. However, explanatory power does not imply predictive power and thus predictive analytics are necessary for assessing predictive power and for building empirical models that predict well. To show that predictive analytics and explanatory statistical modeling are fundamentally disparate, we show that they are different in each step of the modeling process. These differences translate into different final models, so that a pure explanatory statistical model is best tuned for testing causal hypotheses and a pure predictive model is best in terms of predictive power. We convert a well-known explanatory paper on TAM to a predictive context to illustrate these differences and show how predictive analytics can add theoretical and practical value to IS research.
While transportation infrastructure can increase housing price by improving accessibility to opportunities, it generates environmental health risks, such as noise and air pollution, which may have negative effects on housing price. However, the combined effects of accessibility and environmental health risk on housing price have not been well examined in the literature, especially in the auto-oriented urban context of the United States. In this study, we use assessed housing value data and the hedonic model to examine the single-family housing market's reaction to accessibility and environment health risks in Salt Lake County, a growing metropolitan area in Utah experiencing significant air pollution. Three regression models are employed with the consideration of spatial effects: ordinary least squares (OLS), spatial lag regression (SLR), and hierarchical linear modeling (HLM, or multilevel modeling/MLM). By controlling for the influences of structural attributes and socioeconomic conditions, we find that the negative impacts (traffic noise and air pollution) of transportation systems on single-family housing prices are greater than the positive impact (accessibility). Single-family residents in Salt Lake County are willing to pay more to reduce environmental health risks than to get better accessibility. These findings are different from what have been found in some dense and compact urban areas in the literature. These findings suggest that people's willingness to pay for minimizing environmental health risks varies across different urban contexts.
This study identifies the influence of rapid rail systems on property values, with a focus on office buildings in South Africa. It is based on a limited sample of office properties, analysed using MRA to investigate pre-implementation values and rent to post-implementation data. It attempts to confirm if distance from the train stations influences these variables. Evidence is found that distance from the station does have a positive impact on rental levels and property values. The limited data-set, however, causes inadequate levels of statistical significance in some variables, arguably due to the small sample or model specification error due to information availability for research. The positive influence of rapid rail systems found on office values has important implications for property investors, developers, financiers and taxing authorities. This is important amidst a period of extension planning, whereby this research could provide useful information for decision-making and analysis and offers a valuable contribution to the methods to measure the impact of rapid rail systems on property values, although currently limited to office buildings. Furthermore, this research is contributing to the body of knowledge, especially in developing markets, where advanced public transport systems need to be implemented for the first time.
Purpose
The predictive accuracy and reliability of artificial intelligence models, such as the Artificial Neural Network (ANN), has led to its application in property valuation studies. However, a large percentage of such previous studies have focused on the property markets in developed economies and at the same time, effort has not been put into documenting its research trend in the real estate domain. This study nonetheless, critically reviews studies that adopted ANN for property valuation in order to present an application guide for researchers and practitioners and also establish the trend in this research area.
Design/methodology/approach
Relevant articles were retrieved from online databases and search engines and were systematically analysed. Firstly, the background, the construction and the strengths and weaknesses of the technique were highlighted. In addition, the trend in this research area was established in terms of the country of origin of the articles, the year of publication, the affiliations of the authors, the sample size of the data, the number of the variables used to develop the models, the training and testing ratio, the model architecture and the software used to develop the models.
Findings
The analysis of the retrieved articles shows that the first study that applied ANN to property valuation was published in 1991. Thereafter, the technique received more attention from 2000. While a quarter of the articles reviewed emanated from the United States, the rest were conducted in mostly developed countries. Most of the studies were conducted by universities scholars, while very few industry practitioners participated in the researches. Also the accurate predictive ability of the ANN technique was reported in most of the papers reviewed, although a few reported otherwise.
Research limitations/implications
Articles that are not indexed in the search engines and databases searched and also not available in the public domain might not have been captured in this study.
Practical implications
The findings of this study reveals a gap between the valuation practice in developed and developing property markets and also the contributions of real estate practitioners and universities scholars to real estate research. A paradigm shift in the valuation practice in developing nations could lead to achieving a sustainable international valuation practice.
Originality/value
This paper presents the trend in this research area that could be useful to real estate researchers and practitioners in different property markets around the world. The findings of this study could also encourage collaboration between industry professionals and researchers domiciled in both developed and developing countries.
This study uses a hedonic price model to examine the relationship between proximity to newly purchased conservation lands and single-family property values. Specifically, a variant of the hedonic price model is used that addresses changing market values of neighborhood and locational attributes during a given period. Fixed effects are included to control spatial autocorrelation and year and month influences under three variants of an ordinary least-squares (OLS) model, which are double-log, semi-log, and linear model. In addition to the traditional OLS model to explain residential values, a geographically weighted regression (GWR) model is used to study the local difference of coefficient values for each primary variable. An empirical study using a single-family house market-price data set from 2002 to 2010 and 104 newly purchased conservation lands in Alachua County, Florida, is also conducted. To account for the impact of the housing market crash around 2006, the researchers break the data set into two groups (precrash and postcrash), and compare them. The results indicate that sales price increases 0.04% for every percent decrease in distance to the nearest conservation land in general, while the positive influences from conservation lands are larger precrash compared to those afterward. In addition, time from acquisition is not significant precrash; however, it has a negative influence on property values after the housing market crash in 2006. The result is that the purchase of environmentally sensitive lands has an immediate and positive influence on neighboring property values. The influence, however, decreases as time from purchase increases. The results of this study support the government policy of protecting environmentally sensitive lands through their purchase. In addition, coefficient surfaces generated from GWR models can also be used as a guideline for making these purchases in terms of location.
The 2006 sudden and immense downturn in U.S. House Prices sparked the 2007 global financial crisis and revived the interest about forecasting such imminent threats for economic stability. In this paper we propose a novel hybrid forecasting methodology that combines the Ensemble Empirical Mode Decomposition (EEMD) from the field of signal processing with the Support Vector Regression (SVR) methodology that originates from machine learning. We test the forecasting ability of the proposed model against a Random Walk (RW) model, a Bayesian Autoregressive and a Bayesian Vector Autoregressive model. The proposed methodology outperforms all the competing models with half the error of the RW model with and without drift in out-of-sample forecasting. Finally, we argue that this new methodology can be used as an early warning system for forecasting sudden house prices drops with direct policy implications.
Unreliable and inaccurate property valuation has been associated with techniques currently used in property valuation. A possible explanation for these findings may be due to the utilisation of traditional valuation methods. In the current study, an artificial neural network (ANN) is applied in property valuation using the Lagos metropolis property market as a representative case. Property sales transactions data (11 property attributes and property value) were collected from registered real estate firms operating in Lagos, Nigeria. The result shows that the ANN model possesses a good predictive ability, implying that it is suitable and reliable for property valuation. The relative importance analysis conducted on the property attributes revealed that the number of servants’ quarters is the most important attribute affecting property values. The findings suggest that the ANN model could be used as a tool by real estate stakeholders, especially valuers and researchers for property valuation.
Skin cancer, the most common human malignancy, is primarily diagnosed visually, beginning with an initial clinical screening and followed potentially by dermoscopic analysis, a biopsy and histopathological examination. Automated classification of skin lesions using images is a challenging task owing to the fine-grained variability in the appearance of skin lesions. Deep convolutional neural networks (CNNs) show potential for general and highly variable tasks across many fine-grained object categories. Here we demonstrate classification of skin lesions using a single CNN, trained end-to-end from images directly, using only pixels and disease labels as inputs. We train a CNN using a dataset of 129,450 clinical images-two orders of magnitude larger than previous datasets-consisting of 2,032 different diseases. We test its performance against 21 board-certified dermatologists on biopsy-proven clinical images with two critical binary classification use cases: keratinocyte carcinomas versus benign seborrheic keratoses; and malignant melanomas versus benign nevi. The first case represents the identification of the most common cancers, the second represents the identification of the deadliest skin cancer. The CNN achieves performance on par with all tested experts across both tasks, demonstrating an artificial intelligence capable of classifying skin cancer with a level of competence comparable to dermatologists. Outfitted with deep neural networks, mobile devices can potentially extend the reach of dermatologists outside of the clinic. It is projected that 6.3 billion smartphone subscriptions will exist by the year 2021 (ref. 13) and can therefore potentially provide low-cost universal access to vital diagnostic care.
In this paper, we propose a robust methodology to assess the value of microblogging data to forecast stock market variables: returns, volatility and trading volume of diverse indices and portfolios. The methodology uses sentiment and attention indicators extracted from microblogs (a large Twitter dataset is adopted) and survey indices (AAII and II, USMC and Sentix), diverse forms to daily aggregate these indicators, usage of a Kalman Filter to merge microblog and survey sources, a realistic rolling windows evaluation, several Machine Learning methods and the Diebold-Mariano test to validate if the sentiment and attention based predictions are valuable when compared with an autoregressive baseline. We found that Twitter sentiment and posting volume were relevant for the forecasting of returns of S&P 500 index, portfolios of lower market capitalization and some industries. Additionally, KF sentiment was informative for the forecasting of returns. Moreover, Twitter and KF sentiment indicators were useful for the prediction of some survey sentiment indicators. These results confirm the usefulness of microblogging data for financial expert systems, allowing to predict stock market behavior and providing a valuable alternative for existing survey measures with advantages (e.g., fast and cheap creation, daily frequency).
Purpose
The emerging trend in the global real estate valuation practice has led to the development of advanced valuation approaches to replace the traditional methods. The purpose of this paper is to investigate the extent to which real estate valuers practicing in Nigeria are aware and use these advanced approaches in real estate valuation practice.
Design/methodology/approach
Both traditional and advanced approaches were identified from the literature. An online-based questionnaire survey was administered on estate surveyors and valuers to measure their level of awareness and frequency of use of the identified valuation approaches. The feedback was collated and analyzed using descriptive statistical analysis.
Findings
The professionals are mostly aware of the traditional methods and always use the “sales comparison method” in practice. In contrast, they are not very aware of the advanced approaches and hence, only use the hedonic pricing model occasionally in practice.
Research limitations/implications
The study only focuses on the Lagos metropolis, a nationwide survey will produce more comfortable generalizable findings.
Practical implications
This is a wake-up call for the real estate regulatory bodies and indeed all the real estate professionals in Nigeria to embrace the use of the advanced valuation approaches in practice, in order to remain relevant in the international real estate practice.
Originality/value
Implementation of the recommendations of this study could help position the Nigerian real estate professionals and the industry for a global exposition.
Purpose
Real estate property has been established as a composite good, and its value is determined by many variables. The heterogeneous nature of real estate property has made different stakeholders value these variables differently. Therefore, this study aims to identify and evaluate these sets of variables which influence residential property value in the Lagos metropolis property market, Nigeria, based on professional valuers’ perception.
Design/methodology/approach
A list of variables that influences property value was generated through literature review, and the list was used to design an online questionnaire that was administered to valuers practicing in the metropolis. The valuers were asked to rank these variables in order of significance. Their response was analysed to establish the mean score of each variable that depicts their level of significance.
Findings
In order of importance, property location, neighbourhood characteristics, property state of repair, size of property, availability of neighbourhood security and age of property are the most highly significant variables that are influential on the property value in the Lagos metropolis.
Practical implications
The findings of this study will inform all existing and prospective real estate stakeholders, including facility managers of the major determinants of the value of their investments and, at the same time, will be a tool for valuers and researchers in property value modelling.
Originality/value
This study is the first attempt to develop a framework of property value determinants in this research area in Nigeria.
This study takes advantage of the high learning capabilities of data mining (DM) techniques towards to the development of a novel approach for jet grouting (JG) column diameter prediction. The high number of variables involved in JG technology as well as the complex phenomena related with the injection process make JG column diameter (D) prediction a difficult task. Therefore, in order to overcome it, the flexible learning capabilities of DM techniques were applied as an alternative approach of the traditional tools. The achieved results show that both artificial neural network and support vector machine algorithms can be trained to accurately predict D built in different soil types of clayey nature and using different JG systems. In both cases a coefficient of correlation () very close to the unity was achieved. For models training, a set of eight input variables were considered. Among them, the rod withdrawal speed, flow rate of the grout slurry and the JG system were identified as the most relevant ones, although the grout pressure and the dynamic impact of the grout also revealed an important influence on D prediction. Moreover, additionally to the identification of the key model variables, it was also measured their effects on D prediction based on a data-based sensitivity analysis. These achievements represent a novel contribution for JG technology, mainly at the design level. Furthermore, the obtained results also underline the potential and contribution of DM to solving complex problem in geotechnical engineering.
Purpose
– Fluctuations in construction output has an adverse effect on the construction industry and the economy due to its strong linkage. Developing reliable and accurate predictive models is vital to implementing effective response strategies to mitigate the impact of such fluctuations. The purpose of this paper is to compare the accuracy of two univariate forecast models, i.e. Box-Jenkins (autoregressive integrated moving average (ARIMA)) and Neural Network Autoregressive (NNAR).
Design/methodology/approach
– Four quarterly time-series data on the construction output of Hong Kong were collected (1983Q1-2014Q4). The collected data were divided into two parts. The first part was fitted to the model, while the other was used to evaluate the predictive accuracy of the developed models.
Findings
– The NNAR model can provide reliable and accurate forecast of total, private and “others” construction output for the medium term. In addition, the NNAR model outperforms the ARIMA model, in terms of accuracy.
Research limitations/implications
– The applicability of the NNAR model to the construction industry of other countries could be further explored. The main limitation of artificial intelligence models is the lack of explanatory capability.
Practical implications
– The NNAR model could be used as a tool for accurately predicting future patterns in construction output. This is vital for the sustained growth of the construction industry and the economy.
Originality/value
– This is the first study to apply the NNAR model to construction output forecasting research.
This paper uses street-level data on property sales and crime rates for England and Wales to investigate compensating differentials for crime risk. My identification strategy relies on the use of non-parametric regional time trends on various levels of spatial aggregation as well as various fixed effects for streets and wider areas to control for unobserved amenities and regional economic conditions. The data comes from transaction data collected by the land registry and recently published crime maps for the whole of the UK. My estimates, which are robust to a range of sensible specification changes, suggest that each case of anti-social behaviour per ten population in the same street leads to an approximately 0.6–0.8% drop in property prices, while a corresponding increase in violent crime decreases prices by roughly 0.6–1.6% and a corresponding increase in non-violent crime by about 0.2–0.4%. The majority of estimates are at the upper end of these intervals. Estimates for robbery, burglary and vehicle crime are either zero or positive, but are possibly biased because of reverse causality. Crime outside of the respective street does not appear to matter, which is consistent with earlier findings. Expressed in monetary terms each case of anti-social behaviour costs society between £5000 and £6700 and each violent crime between £5000 and £13,300. The results confirm estimates based on prior willingness-to-pay studies and other studies using smaller areas such as single cities.
Purpose
– The purpose of this paper is to review the issues involved in the implementation of mass valuation systems and the conditions needed for doing so.
Design/methodology/approach
– The method makes use of case studies of and fieldwork in countries that have either recently introduced mass valuations, brought about major changes in their systems or have been working towards introducing mass valuations.
Findings
– Mass valuation depends upon a degree of development and transparency in property markets and an institutional structure capable of collecting and maintaining up-to-date price data and attributes of properties. Countries introducing mass valuation may need to undertake work on improving the institutional basis for this as a pre-condition for successful implementation of mass valuation.
Practical implications
– Although much of the literature is concerned with how to improve the statistical modelling of market prices, there are significant issues concerned with the type and quality of the data used in mass valuation models and the requirements for successful use of mass valuations.
Originality/value
– Much of the literature on mass valuation takes the form of the development of statistical models of value. There has been much less attention given to the issues involved in the implementation of mass valuation.
We examined the effects of refinery air pollution on house prices near Houston, Texas. The affected area was identified through AERMOD air modeling of past releases of sulfur dioxide, a proxy for respiratory risk. A total of 3,964 residential MLS sales from 2006 to 2011 were used to populate an OLS model, a spatial model, and a spatial model with an additional endogenous variable. The findings indicate that air pollution has a significant negative 6%-8% loss on house prices. For one year, the negative effect is shown to generally diminish with distance up to about two miles from the refinery.
Corrosion is often a major failure mechanism for concrete sewers and under such circumstances the sewer service life is largely determined by the progression of microbially induced concrete corrosion. The modelling of sewer processes has become possible due to the improved understanding of in-sewer transformation. Recent systematic studies about the correlation between the corrosion processes and sewer environment factors should be utilized to improve the prediction capability of service life by sewer models. This paper presents an artificial neural network (ANN)-based approach for modeling the concrete corrosion processes in sewers. The approach included predicting the time for the corrosion to initiate and then predicting the corrosion rate after the initiation period. The ANN model was trained and validated with long-term (4.5 years) corrosion data obtained in laboratory corrosion chambers, and further verified with field measurements in real sewers across Australia. The trained model estimated the corrosion initiation time and corrosion rates very close to those measured in Australian sewers. The ANN model performed better than a multiple regression model also developed on the same dataset. Additionally, the ANN model can serve as a prediction framework for sewer service life, which can be progressively improved and expanded by including corrosion rates measured in different sewer conditions. Furthermore, the proposed methodology holds promise to facilitate the construction of analytical models associated with corrosion processes of concrete sewers.
In this chapter we consider bounds on the rate of uniform convergence. We consider upper bounds (there exist lower bounds as well (Vapnik and Chervonenkis, 1974); however, they are not as important for controlling the learning processes as the upper bounds).
In recent years, social, economic and fiscal factors have produced strong modifications of the Italian real estate market, that currently appears as a complex system characterized by continuous transformation. In this context, for real estate operators the use of “slender” tools, able i) to operate even on limited data, ii) to automatically capture from data the causal relations between explanatory variables and prices, iii) to appraisal the market values that will reasonably occur in the short term, has become essential. Among artificial intelligence models, artificial neural networks (ANN) meet these prerogatives. In this paper, ANN is applied to the evaluation of market values of residential properties starting from a sample of apartments recently sold in a neighborhood of the city of Bari (Italy). The excellent results confirm the effectiveness of ANN in property valuations. The work must be attributed in equal parts to the authors.
The advancement of computational software within the last decade has facilitated enhanced uptake of mass appraisal methodologies by the valuation and prediction accuracy in computer-assisted mass appraisal community for price modelling, estimation and tribunal defence. Applying a sample of 2694 residential properties, this paper assesses and analyses a number of geostatistical approaches relative to an artificial neural network (ANN) model and the traditional linear hedonic pricing model for mass appraisal valuation accuracy and price estimation purposes. The findings demonstrate that the geostatistical localised regression approach is superior in terms of model explanation, reliability and accuracy. ANNs can be shown to perform very well in terms of predictive power, and therefore valuation accuracy, outperforming the traditional multiple regression analysis (MRA) and approaching the performance of spatially weighted regression approaches. However, ANNs retain a ‘black box’ architecture that limits their usefulness to practitioners in the field. In relation to cost-effectiveness and user-friendly applicability for the valuation community, the MRA approach outperforms the ‘black box’ nature of the ANN technique, with the geographically weighted regression approach providing the best balance of outright performance and transparency of methodology. It is this spatially weighted approach utilising absolute location which appears to represent the way forward in developing the practice of mass appraisal.
Economic Modeling Using Artificial Intelligence Methods examines the application of artificial intelligence methods to model economic data. Traditionally, economic modeling has been modeled in the linear domain where the principles of superposition are valid. The application of artificial intelligence for economic modeling allows for a flexible multi-order non-linear modeling. In addition, game theory has largely been applied in economic modeling. However, the inherent limitation of game theory when dealing with many player games encourages the use of multi-agent systems for modeling economic phenomena. The artificial intelligence techniques used to model economic data include: multi-layer perceptron neural networksradial basis functionssupport vector machinesrough setsgenetic algorithmparticle swarm optimizationsimulated annealingmulti-agent systemincremental learningfuzzy networksSignal processing techniques are explored to analyze economic data, and these techniques are the time domain methods, time-frequency domain methods and fractals dimension approaches. Interesting economic problems such as causality versus correlation, simulating the stock market, modeling and controling inflation, option pricing, modeling economic growth as well as portfolio optimization are examined. The relationship between economic dependency and interstate conflict is explored, and knowledge on how economics is useful to foster peace and vice versa is investigated. Economic Modeling Using Artificial Intelligence Methods deals with the issue of causality in the non-linear domain and applies the automatic relevance determination, the evidence framework, Bayesian approach and Granger causality to understand causality and correlation. Economic Modeling Using Artificial Intelligence Methods makes an important contribution to the area of econometrics, and is a valuable source of reference for graduate students, researchers and financial practitioners.