No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Forecasting U.S. Tourist arrivals using optimal Singular Spectrum Analysis
... The results were similar for short-term periods; however, SSA had a better performance at longer horizons. Hassani et al. (2015) analyzed the advantages of using SSA compared with ARIMA, exponential smoothing (ETS), and NN. The authors indicated the superiority of the SSA model over other single models in forecasting tourist arrivals into the US. ...
... Therefore, to avoid this problem and ensure that no future information is taken to the filtered time series, we choose the L and r based on the best performance of the combination of the parameters obtained in the training step. In our study, we follow the procedure highlighted by Hassani et al. (2015). However, it is worth mentioning that to predict the parameters L and r, we apply the same forecasting models in the out-of-sample step rather than use the SSA algorithm described in the Hassani et al. (2015) procedure. ...
... In our study, we follow the procedure highlighted by Hassani et al. (2015). However, it is worth mentioning that to predict the parameters L and r, we apply the same forecasting models in the out-of-sample step rather than use the SSA algorithm described in the Hassani et al. (2015) procedure. ...
Paper aims: To predict monthly corn, soybean, and sugar spot prices in Brazil using hybrid forecasting techniques.
Originality: This study combines the Singular Spectrum Analysis with different forecasting methods.
Research method: This paper presents a set of hybrid forecasting approaches combining Singular Spectrum Analysis(SSA) with different univariate time series methods, ranging from complex seasonality methods to machine learning and autoregressive models to predict monthly corn, soybean, and sugar spot prices in Brazil. We carry out a range of out-of-sample forecasting experiments and use a comprehensive set of forecast evaluation metrics. We contrast the performance of the proposed approaches with that of a range of benchmark models.
Main findings: The results show that the proposed hybrid models perform better, with the hybrid SSA-neural network approach providing the most competitive results in our sample.
Implications for theory and practice: Forecasting agricultural prices is of paramount importance to assist producers, farmers, and the industry in decision-making processes.
... Furthermore, exponential smoothing (ES) models are less commonly cited in the research on tourism demand volatility, notwithstanding the overwhelming evidence that ES frequently produce reasonable forecasting of orientation and trend shifts in tourism demand [3,14,54,84,85]. Just like Cho (2003) in [54], Coshall (2009) in [9], Coshall and Charlesworth (2011) in [37], and Lim and McAleer (2001) in [86] analyzed the performance of the ES method in forecasting tourist arrivals. ...
... Just like Cho (2003) in [54], Coshall (2009) in [9], Coshall and Charlesworth (2011) in [37], and Lim and McAleer (2001) in [86] analyzed the performance of the ES method in forecasting tourist arrivals. Especially, Hassani et al. (2015) in [84] contrasted singular spectrum analysis with ARIMA, exponential smoothing (ETS), and neural networks (NN) for tourism demand forecasting. Xu et al. (2016) in [87] employed Naïve 1, simple exponential smoothing (SES), double exponential smoothing (DES), ARIMA, and other approaches to estimate tourist arrivals in Hong Kong (China). ...
... Just like Cho (2003) in [54], Coshall (2009) in [9], Coshall and Charlesworth (2011) in [37], and Lim and McAleer (2001) in [86] analyzed the performance of the ES method in forecasting tourist arrivals. Especially, Hassani et al. (2015) in [84] contrasted singular spectrum analysis with ARIMA, exponential smoothing (ETS), and neural networks (NN) for tourism demand forecasting. Xu et al. (2016) in [87] employed Naïve 1, simple exponential smoothing (SES), double exponential smoothing (DES), ARIMA, and other approaches to estimate tourist arrivals in Hong Kong (China). ...
Tourism forecasting has garnered considerable interest. However, integrating tourism forecasting with volatility is significantly less typical. This study investigates the performance of both the single models and their combinations for forecasting the volatility of tourism demand. The seasonal autoregressive integrated moving average (SARIMA) model is used to constr0uct the mean equation, and three single models, namely the generalized autoregressive conditional heteroscedasticity (GARCH) family models, the error-trend-seasonal exponential smoothing (ETS-ES) model, and the innovative smooth transition exponential smoothing (STES) model, are employed to estimate the volatility of monthly tourist arrivals into Malaysia. This study also assesses the accuracy of forecasts using simple average (SA), minimum variance (MV), and novel smooth transition (ST). STES performed the best of the single models for forecasting the out-of-sample of tourism demand volatility, followed closely by ETS-ES. In contrast, the ST combining method surpasses SA and MV. Interestingly, forecast combining methods do not always outperform the best single model, but they consistently outperform the worst single model. The MCS and DM tests confirm the aforementioned findings. This article merits consideration for future forecasting research on tourism demand volatility.
... As for the filtering methods, they commonly decomposed the observed variable into trend and cycle by filtering, such as Hodrick-Prescott filter and Hamilton filter (Bosupeng 2019). As for the denoising methods, the mainly involved step is to remove the time series noise and then use the reconstructed time series with the forecasting techniques for prediction, such as Singular-Spectrum Analysis (SSA) (Hassani et al. 2015) and wavelet decomposition (Kummong and Supratid 2016). As for the multiscale methods, the idea is to regard time series as the superposition of different frequency components, which usually include seasonal trend decomposition based on Loess (STL) (Y. ...
... In particular, Kummong and Supratid (2016) proposed a hybrid forecasting model with the discrete wavelet decomposition leading to accurate performance, by extracting the hidden features of Thailand's tourism time series at various frequency resolution levels. Hassani et al. (2015) employed SSA for forecasting the monthly US tourist arrivals by filtering the noise and achieved a better prediction accuracy. Moreover, Hassani et al. (2017) also used the SSA for denoising the monthly international tourist arrivals in European countries and obtained similar higher performance. ...
... Evaluation criteria. To ensure consistency with evaluation measures of the previous tourism forecasting studies, the mean absolute percentage error (MAPE), root mean square error (RMSE), the improvement ratio (IR), and the directional accuracy (DA) are used to evaluate the performance, in which MAPE and RMSE are the level predictions criteria and the DA is the directional prediction criteria(Hassani et al. 2015;Sun et al. 2019;X. Li and Law 2020): Tourist arrivals to Hong Kong from eight major markets. ...
Decomposition methods are extensively used for processing the complex patterns of tourism demand data. Given tourism demand data’s intrinsic complexity, it is critical to theoretically understand how different decomposition methods provide solutions. However, a comprehensive comparison of decomposition methods in tourism demand forecasting is still lacking. Hence, this study systematically investigates the forecasting performance of decomposition methods in tourism demand. Nine popular decomposition methods and six forecasting methods are employed, and their forecasting performance is compared. With Hong Kong visitor arrivals from eight major sources as a sample, three main conclusions are obtained from empirical results. First, all the decomposition methods generally outperform benchmark at all horizons, in both the level and directional forecasting. Second, decomposition methods can be divided into four categories based on forecasting accuracy. Finally, variational mode decomposition method is consistently superior to other eight decomposition methods and can provide the best forecasts in all cases.
... They assert that the results of the proposed method are effective in terms of increasing accuracy for forecasting. Hassani et al. [41] analyze the results of singular spectrum analysis (SSA), ARIMA, exponential smoothing (ETS), and neural networks (NN), and the performance comparison shows the higher performance of SSA among other models. With the aim of obtaining a better performance, Cuhadar [42] compares exponential smoothing, Box-Jenkins, and an ANN model for Turkey's tourism revenues forecasting and concludes that the ANN model outperforms the other models. ...
... In our analysis, except for hybrid models among the top five models, the forecasting power of ETS and the grey model is found to have overperformance compared to other models. Comparing to the literature, although the performance of the ETS model for forecasting is seen as lower than the alternative models [41,42], our findings support the opposite. On the other hand, the grey model is generally addressed in the literature by optimizing it or with another model to prevent the disadvantages of this model [55,56,94]. ...
Accurate forecasting of tourism demand and income holds paramount importance for both the tourism industry and the national economy. This study aims to address several objectives: (1) specify the best forecasting model in the prediction of tourist arrival volumes and tourism income for Turkey; (2) assess the degree of impact exerted by various determinants on the tourism forecasts; (3) generate forecasts for tourist arrival volumes and tourism income using the most suitable models; and (4) examine potential scenarios illustrating the ramifications of the Russia-Ukraine war on tourist arrival volumes and tourism income. The forecasting models employed in this study encompass a comprehensive set of statistical methods, including ETS, ARIMA, TRAMO-SEATS, X13, X11, STL, Grey, and their combinations with ANN. In the ANN models, exogenous variables such as the global financial crisis, the Turkey-Russia warplane crash crisis, the COVID-19 pandemic, and USD/TRY exchange rates are incorporated. The results unveil the identification of five superior models: ETS, Grey, hybrid ETS-ANN, hybrid Grey-ANN, and hybrid ARIMA-ANN models, which exhibit the lowest MAPE and sMAPE values. Forecasts for the forthcoming quarters are examined under two scenarios: assuming the continuity or cessation of the Russia-Ukraine war. Comparative analysis of the relative effects of exogenous variables indicates that COVID-19 has the most substantial impact on tourist arrival volumes, and tourism income is primarily influenced by the USD/TRY exchange rate.
... The research work of [23] compared SSA forecasting ability of SSA, ARIMA, ES, and feed-forward Neural Network model (NN) using monthly U.S. Tourist arrivals from 1996 to 2012. The authors employed cross-validation where two-thirds of the sample data (n1 = 355) formed the training set (used to tune the model parameters), and a third (n2 = 356 observations) formed the validation set (used to test the forecast accuracy over a given horizon). ...
... The results are coherent with empirical studies such as those of [11], who found a gradual increase in RRMSE with an increase in L in reconstructing a simulated sinusoidal noisy series. Using a sample of 355 observations of monthly US Tourist arrivals data as the training set, [23] found that a window length of 15 with associated 6 eigenvectors was the best-fit window length minimization of the prediction errors (MAPE = 9%). Studies that have used a relatively larger window length have also found few significant eigentriples to reconstruct the series, implying that a higher dimensional matrix is unnecessary as it reflects data redundancy -a possible cause of overfitting data. ...
Time series modeling and forecasting techniques serve as gauging tools to understand the time-related properties of a given time series and its future course. Most financial and economic time series data do not meet the restrictive assumptions of normality, linearity, and stationarity of the observed data, limiting the application of classical models without data transformation. As non-parametric methods, Singular Spectrum Analysis (SSA) is data-adaptive; hence do not necessarily consider these restrictive assumptions as in classical methods. The current study employed a longitudinal research design to evaluate how SSA fist Kenya’s monthly industrial inputs price index from January 1992 to April 2022. Since 2018, reducing the costs of industrial inputs has been one of Kenya’s manufacturing agendas to level the playing field and foster Kenya’s manufacturing sector. It was expected that Kenya’s Manufacturing Value Added hit a tune of 22% by 2022. The study results showed that the SSA (L = 12, r =7) (MAPE = 0.707%) provides more reliable forecasts. The 24-period forecasts showed that the industrial inputs price index remains high above the index in 2017 before the post-industrial agenda targeting a reduction in the cost of industrial inputs. Thus, the industrial input prices should be reduced to a sustainable level.
... Theoretical analysis is needed to justify if such processing is suitable in obtaining the optimum forecasting outcome. Apart from raw data characteristics, data decomposition can also help improve the forecasting accuracy (Chan et al., 2005; H. Hassani et al., 2015;X. Li & Law, 2020;Hassani et al., 2017). ...
... X. Li and Law (2020) claimed that the empirical model decomposition could be used to generate stationary series and accurate forecasting performance. Along the same line of research, Hassani et al. (2015) used Singular Spectrum Analysis (SSA) decomposition in tourism development analysis with US tourism demand forecasting. Y. Zhang, Li, Muskat, Law (2020) concluded that the Seasonal Trend Loses (STL) decomposition with deep learning could improve the tourism demand forecasting. ...
As tourism researchers continue to search for solutions to determine the best possible forecasting performance, it is important to understand the maximum predictivity achieved by models, as well as how various data characteristics influence the maximum predictivity. Drawing on information theory, the predictivity of tourism demand data is quantitatively evaluated and beneficial for improving the performance of tourism demand forecasting. Empirical results from Hong Kong tourism demand data show that 1) the predictivity could largely help the researchers estimate the best possible forecasting performance and understand the influence of various data characteristics on the forecasting performance.; 2) the predictivity can be used to assess the short effect of external shock — such as SARS over tourism demand forecasting.
... Models are known for filtering capabilities in any time series, the possibility to enrich analysis by using SSA, and to choose the window length and the number of eigenvalues in the MSSA model (Silva et al., 2018). Algorithms that improved an SSA model with the loss function and made the model more universal to any forecasting issue were presented (Hassani et al., 2015). SSA was outperformed by MSSA in the study of Silva et al. (2018), but there is no single model suitable for every element or forecast of the chosen country's performance. ...
... Optimal failure repair time must be within 40 minutes to achieve 90% maintainability(Tsarouhas & Arvanitoyannis, 2012) Operational excellence profitability (OEP) indicators. The OEP indicators that were related to performance time losses (energy consumption, direct labor used) showed cost benefits (in total $1564,27), and those that were related to quality (raw materials loss, packaging materials loss, maintenance labor extra time, maintenance spare parts) were not beneficial (Gólcher-Barguil et al., 2019) Qualitative data utilization for industrial performance assessment. Qualitative information indicated a higher level of project goal achievement from the level of 0.68 to 0.92. ...
Over the years, technological progress has accelerated highly, and the speed, flexibility, human error reduction, and the ability to manage the process in real time have become more critical and required production companies to adapt production and business models according to the needs. The demand for real-time decision support systems adapted to these raising business needs is continuously growing. Nevertheless, businesses usually face challenges in identifying new indicators, data sources, and appropriate financial modeling methods to analyze them. This paper aims to define and summarize the main financial/economic forecasting methods for production companies in the context of Industry 4.0. Main findings show forecasting accuracy of up to 96% when combining economic and demand information, optimal forecasting period from 10 months to five years, more frequent use of soft indicators in forecasting, the relationship between company’s size and production planning. Four groups of indicators used in financial modeling, such as (I) production-related, (II) customers’ and demand-oriented, (III) industry-specific, and (IV) media information indicators, were separated. The analysis forms a suggestion for decision-makers to pay more attention to the forecasting object identification, indicators’ selection peculiarities, data collection possibilities, and the choice of appropriate methods of financial modeling.
AcknowledgmentThis work was partly supported by Project No. 0121U100470 “Sustainable development and resource security: from disruptive technologies to digital transformation of Ukrainian economy”.
... One method that can handle structural breaks and optimize the presence of frequency components in the data is SSA. A preliminary study has been carried out on the use of structural-break modeling with SSA by [37,38,[40][41][42][43]. The results showed that in modeling structural breaks, the linear recurrent formulae (LRF) form recurrent equations in the SSA framework, leading to a biased conclusion. ...
Standard time-series modeling requires the stability of model parameters over time. The instability of model parameters is often caused by structural breaks, leading to the formation of nonlinear models. A state-dependent model (SDM) is a more general and flexible scheme in nonlinear modeling. On the other hand, time-series data often exhibit multiple frequency components, such as trends, seasonality, cycles, and noise. These frequency components can be optimized in forecasting using Singular Spectrum Analysis (SSA). Furthermore, the two most widely used approaches in SSA are Linear Recurrent Formula (SSAR) and Vector (SSAV). SSAV has better accuracy and robustness than SSAR, especially in handling structural breaks. Therefore, this research proposes modeling the SSAV coefficient with an SDM approach to take structural breaks called SDM-SSAV. SDM recursively updates the SSAV coefficient to adapt over time and between states using an Extended Kalman Filter (EKF). Empirical results with Indonesian Export data and simulation studies show that the accuracy of SDM-SSAV outperforms SSAR, SSAV, SDM-SSAR, hybrid ARIMA-LSTM, and VARI.
... In the first stage, we decompose the series to enable signal extraction and noise reduction. In the second stage, we reconstruct a less noisy series and use the reconstructed series for forecasting new data points [1]. In the MSSA method there are two forms for the trajectory matrix: vertical and horizontal. ...
In time series analysis, ignoring outliers leads to misidentification of the model, biased estimation of parameters, and poor predictions. One of the reliable non-parametric methods in predicting and improving the quality of multivariate time series modeling is the multivariate Singular Spectrum Analysis (MSSA) technique, which does not require any initial assumptions. The presence of outliers affects the Frobenius matrix norm and reduces the MSSA method's efficiency. This research proposes a new version of MSSA based on the L1- norm. Then the performance of this method is compared with basic MSSA using simulation studies and real data.
... Tanto no SSA quanto no MSSA, a decisão mais importanteé sobre a escolha dos parâmetros L e m. A abordagem usada mais recentemente para tal tarefaé descrita em Hassani et al. [2015]. Em termos gerais: (1) a sérieé dividida em treinamento e validação; (2) uma gama de combinações possíveis de parâmetrosé empregada na parte de treinamento; (3) para cada combinação, as previsões são obtidas; (4) calcula-se o erro entre a previsão e os valores observados para cada conjunto de parâmetros e decidi-se pelo conjunto de parâmetros que obteve os menores erros. ...
... SSA has received a great deal of attention in the prediction field recently. It was used to forecast required staffing levels (Gillard & Knight, 2014), tourism demand, daily pound/dollar exchange rate, economic and financial time series, the industrial production economies and death series (Hassani, 2007;Hassani et al., 2009;Hassani & Thomakos, 2010;Hassani et al., 2015). The SSA-based model outperforms Automatic-ARIMA, Neural Networks methods and Holt-Winters' methods. ...
Accepted by: Konstantinos Nikolopoulos
One of the challenges of emergency ambulance demand (EAD) time series prediction lies in their non-stationary nature. We study this important problem and propose two hybrid forecasting models, which combine the singular spectrum analysis (SSA) time-series technique with autoregressive integrated moving average (ARIMA) parameterized multivariate forecasting. Both daily and hourly time series are studied. The non-stationary time series are decomposed into three eigentriples by SSA: trends, periodic components and residuals. Selection of the group boundary point of the periodic component is a key issue in the SSA method. We use spectrum analysis to compute a threshold for maximum information content of periodic components. ARIMA mean value prediction models are employed to forecast the trends, periodic components and residuals sub-series. Our research compares ARIMA and SSA-based hybrid models by considering the emergency dispatching departure records of six core districts in Guangzhou city from 1 January 2021 to 31 December 2021. Results show that the integrated SSA-ARIMA model performs best. SSA is a very effective pre-processing method for non-stationary time series prediction. The predictive accuracy of using a hybrid model for hourly EAD time series is higher than that for daily ones. Our discussion should be useful for improving EAD prediction in contexts others than that considered in our research.
... The DOSM website provided the monthly electricity usage figures for domestic use from January 2018 to April 2022 that were used in this study. As stated by Hamzah et al. (2018) and Hassani et al. (2015) data is split into two part which are training and testing data to evaluate forecast error using genuine forecasts. This procedure aims to give a trustworthy estimate of how well both types of data are suited by the model in accordance with Hyndman and Athanasopoulos (2018). ...
... These seasonal variations in arrivals, as proxies of tourism demand, can cause a temporary mismatch between the supply and demand of tourism facilities. Thus, business owners must try to manage this mismatch by creating temporary employment or developing marketing strategies based on monitoring short-term information about demand, including promotional campaigns or dynamic pricing (Lim and Mcaleer, 2001; Bartolomé et al., 2009;Hadavandi et al., 2011;Song et al., 2011;Hassani et al., 2015Hassani et al., , 2017García Rodríguez, 2017;Thushara et al., 2019;Vergori and Arima, 2022;Xing et al., 2022). Short-term variations are particularly relevant to airport managers to optimise their resources to adjust capacity to the current demand and avoid congestion, as well as dealing with peaks, preparing infrastructure and spaces and improving connectivity (Volner and Smrz, 2009;Beria and Laurino, 2016;Aygün et al., 2017;Mendes Dantas et al., 2017;Grbcic et al., 2021;Long et al., 2021;Lunacek et al., 2021;Sismanidou et al., 2022). ...
The aim of this paper is to describe and compare seasonal effects in daily air passenger arrivals. Multiple seasonal cycles of different lengths are usually observed in daily time series. To model weekly cycles over a short and fixed seasonal period, conventional formulations can be applied. However, the length of yearly and, above all, monthly seasonal cycles is not fixed. Yet, whatever the lengths of the seasonal periods, seasonal cycles can be modelled by using evolving spline functions in such a way that a seasonal effect at a proportion of the seasonal period is defined as a non-fixed parametric formulation of this proportion. Thus, restricted evolving spline models are useful to model this type of seasonal variation. The areas under the splines can measure changes in the
magnitude of seasonal variations over time and can be used to compare the relevance of seasonal variations for different seasonal periods or different time series. This proposal is illustrated by applying it to daily series of air passengers arriving in the Canary Islands and at the rest of Spain’s airports.
... Meanwhile, exponential smoothing (ETS) models are used less frequently in tourism demand studies, although there is ample evidence that ES often makes reasonable predictions of shifts in the direction and trend of tourism demand. Similarly, in tourism demand forecasting, the performance of singular spectrum analysis was compared to ARIMA, exponential smoothing (ETS), and neural networks (NN) [8]. After using Nave 1, simple exponential smoothing (SES), double exponential smoothing (DES), ARIMA, and other methods to estimate tourist arrivals in Hong Kong [9], Annual international tourist arrivals in Zambia were studied using ARIMA and Holt-Winters exponential smoothing [10]. ...
As a result of the 2019 epidemic, all industries around the world have been hit hard, especially the tourism industry. Governments have introduced many measures to maintain the tourism industry, but the results are not satisfactory, and the industry has even experienced negative growth. In order to predict the future trend of the tourism industry and to make decisions, this paper has selected five different data on airlines, hotels, car rentals and travel agencies in the US tourism industry as the basis, and will use two models, ETS and ARIMA, to forecast the data from 2000 to 2020 respectively, to obtain the data without the impact of the epidemic, and then compare the forecast results with the historical real data. The results will then be compared with the historical data. Finally, a model suitable for each data was obtained.
... Presumably, those effects change over time, as society becomes increasingly concerned with environmental issues. Short-and long-term patterns of tourism demand have been addressed with univariate and bivariate decomposition frameworks, extensively being used to isolate trends and cycles (Hassani et al., 2015;Li & Law, 2019;or Ridderstaat & Croes, 2017). These trends or tendencies, or the lack thereof, are relevant for many stakeholders in the tourism industry, as being beyond their influence. ...
An analytical review of the literature on the effect of the Blue Flag eco-label on tourism demand reveals clear divergences. Certain free sources provide valuable panel data to study trends over time in Blue Flag awards to beaches, on one side, and the demand for accommodation in bathing season, on the other. These data also serve as a basis for the construction of correlation models, such as the simple approach tested here in a representative territory of sun&beach tourism, on the south and east coasts of Spain. As a result, these two variables follow a similar trend, and the Blue Flag seems to influence the inflows of both foreign and domestic tourists to that destination. Nonetheless, for the sake of consistency, factors other than Blue Flag should be considered, and the study period extended.
... Penelitian terdahulu oleh (Chaari et al., 2012) yang meneliti kadar IGF-1 pada anak usia laki-laki usia 10-12 terdapat perbedaan nilai kadar IGF-1 yang signifikan antara kelompok permainan bola voli intensitas rendah dengan kelompok kontrol yang tidak diberi perlakuan. (Hassani et al., 2015) meneliti kelompok eksperimen dan kelompok kontrol (tidak diberi perlakuan) dengan jumlah subjek anak usia 10 tahun dengan metode respons intervensi olahraga renang dengan intensitas 70-80% dengan hasil terdapat pengaruh yang signifikan terhadap GH (Growth Hormone), namun tidak terdapat pengaruh yang signifikan pada hormon IGF-1. Pada penelitian sebelumnya menunjukkan bahwa olahraga jalan kaki yang intens selama 6 minggu dapat memengaruhi kadar serum IGF-1 dan IGFBP3 serta bio-marker metabolik termasuk kolesterol kepadatan tinggi, glukosa, dan trigliserida (Kim et al., 2015). ...
Tujuan pada penelitian ini untuk mengetahui pengaruh permainan bola voli yang dimodifikasi terhadap kadar hormon Insuline-like growth factor-1. Metode yang digunakan The Randomized Pretest-Posttest Group Design dengan Pengambilan sampel menggunakan teknik rumus oleh Higgins dan Kleinbaum, dengan hasil 20 subjek. Selanjutnya subjek ditentukan berdasarkan kriteria inklusi dan eksklusi dan di bagi kedalam kelompok eksperimen dan kontrol. Kelompok eksperimen diberikan intervensi modifikasi permainan bola voli dan kelompok kontrol tanpa perlakuan. Darah diambil 24 jam sebelum intervensi olahraga dan 24 jam setelah intervensi. Modifikasi permainan bola voli dilakukan selama 6 minggu dengan 3x intervensi dalam 1 minggu. Hasil data serum di analisis dengan SPSS 25. Hasil dari uji One Way Anova kelompok modifikasi dengan kelompok control menunjukkan nilai p value yaitu 0,000. Sedangkan pada uji Paired Sample T-Test untuk tes awal dan tes akhir pada kelompok modifikasi permainan bola voli nilai p value yaitu 0,000. Hasil penelitian ini disimpulkan ada pengaruh menggunakan modifikasi permainan bola voli terhadap anak prepubertas.
... Moreover, a large amount of prominent research has proven that SSA can effectively extract the features of original time series, i.e., trend components, periodic components, and noise components, which is helpful in identifying the different components contained in an original signal. For example, Hassani et al. (2015) proposed an ensemble model based on SSA for tourist demanding forecasting, and the results showed that it has a competitive advantage over other popular methods. Later, Lahmiri (2018) established an ensemble model based on SSA and SVR, and Wang, Zhang, and Ma (2020) utilized SSA and the Laguerre neural network to establish a hybrid ensemble forecasting model. ...
Forecasting complex time series faces a huge challenge due to its high volatility. To improve the accuracy and robustness of prediction, this paper proposes a bi‐level ensemble learning approach by combining decomposition ensemble forecasting and resample strategies. The bi‐level ensemble approach consists of four steps: data decomposition via singular spectrum analysis (SSA), resampling by employing a bagging algorithm, individual forecasting utilizing extreme learning machine (ELM) and introducing sorting pruning and simple addition ensemble strategies for integrating the inner and outer results, respectively. To verify the effectiveness of the established forecasting approach, three exchange rate time series are selected as samples. The results reveal that the proposed model is significantly better than the other benchmarks at different lead times, which indicates that it can be regarded as an effective and promising tool for complex time series forecasting.
... It basically decomposes the time series into a sum of independent and interpretable components and also provides information about each additive component (Hassani, 2007). For the last two decades, SSA has been applied over vast data fields, for example, in nonlinear dynamics geoscience and economics studies (Broomhead & King, 1986;Ghil & Vautard, 1991;Ghil et al., 2002;Hassani et al., 2014;Keppenne & Ghil, 1990;Kikuchi & Wang, 2008;Lisi & Medio, 1997;Schoellhamer, 2001;Vautard & Ghil, 1989;Vautard et al., 1992). This method is a very constructive tool for (1) extracting trends, (2) extracting seasonality components and cycles of periodicity, (3) smoothing and separating noise from signal, (4) separating periodicities with varying amplitudes, (5) finding the structure of a short time series, and (6) detection of change points (Hassani, 2007). ...
Radon-222 is considered an important earthquake precursor because it is a noble gas and for its radioactive decay characteristics. However, radon anomalies are controlled not only by seismic activity, but also by environmental and other related natural phenomena such as atmospheric pressure, atmospheric temperature, and Earth tides. These effects must be assessed and eliminated from the radon signal to enhance its capability as a precursor. To identify these effects and their modes of variability in radon progeny time series, we have applied singular spectrum analysis (SSA). We analyzed hourly measurements of radon progeny for a 6-month period acquired from a 10-m-deep borehole at the Multi-Parametric Geophysical Observatory at Ghuttu, Garhwal Himalaya. The site also has a superconducting gravimeter for acquiring continuous gravity suitable to detect tidal waves with high accuracy. Our results show clear evidence of diurnal and semidiurnal daily periodicities, along with slight signatures of the M2–O1 waves in the radon time signal. Results show that even at 10 m depth these effects play a part in controlling radon release from the soil.
... In the above neural network with a hidden layer, logit function provides a non-linear fit. K represents the nodes in the hidden layer 21 . It is worth mentioning that the activation function in this type of applied ANN was a combination of logit link and autoregressive models. ...
Providing fresh blood to keep people in need of blood alive, has always been a main issues of health systems. Right policy-making in this area requires accurate forecasting of blood demand. The current study aimed at predicting demand for different blood groups in Shiraz using Auto Regressive Integrated Moving Average (ARIMA), Artificial Neural Network (ANN) and a hybrid approaches. In the current time series analysis, monthly data of the Shiraz hospitals and medical centers demand for 8 blood groups during 2012–2019 were gathered from Shiraz branch of Iranian Blood Transfusion Organization. ARIMA, ANN and a hybrid model of them was used for prediction. To validate and comprise ARIMA and ANN models, Mean Square Error (MSE) and Mean Absolute Error (MAE) criteria were used. Finally, ARIMA, ANN and hybrid model estimates were compared to actual data for the last 12 months. R3.6.3 were used for statistical analysis. Based on the MSE and MAE of models, ARIMA had the best prediction for demand of all blood groups except O+ and O−. Moreover, for most blood groups, ARIMA had closer prediction to actual data. The demand for four blood groups (mostly negative groups) was increasing and the demand for other four blood groups (mostly positive ones) was decreasing. All three approaches including ARIMA, ANN and the hybrid of them predicted an almost downward trend for the total blood demand. Differences in the performance of various models could be due to the reasons such as different forecast horizons, daily/month/annual data, different sample sizes, types of demand variables and the transformation applied on them, and finally different blood demand behaviors in communities. Advances in surgical techniques, fetal screening, reduction of accidents leading to heavy bleeding, and the modified pattern of blood request for surgeries appeared to have been effective in reducing the demand trend in the current study. However, a longer time period would certainly provide more accurate estimates.
... The potential of using Singular Spectrum Analysis (SSA) was examined by Hassani, Webster, Silva and Heravi (2015) using tourist arrivals into United States of America. These authors found that SSA offers significant advantages than alternatives methods, like Autoregressive Integrated Moving Average, exponential smoothing and neural networks. ...
The main objectives of this research are to identify, through a systematic literature review, the potential benefits of the use of volatility models in tourism, to study the volatility of tourism demand in cities and to compare models of volatility between different destinations and source markets. The three cities analysed in Portugal were Coimbra, Lisbon and Oporto and the source markets that were studied were the domestic market, the total overnight stays, Brazil, France, Germany, Italy, Spain, the United Kingdom and other non-specified countries.
The systematic review of the literature was carried out in order to identify, in a temporal
perspective, the use of each methodology, variables used, data frequencies, temporal window, type of territories and geographic object of each study. The semantic analysis of the state of the art was also a methodology used. After a preliminary analysis of the time series, models that literature indicates as more suitable to estimate the volatility were used, namely, models of autoregressive conditional heteroscedasticity: ARCH, GARCH, EGARCH and TGARCH models.
The most suitable models for each source market, in each city, were identified, as well as the existence of asymmetries face to positive and negative shocks, their magnitude and their persistence. Different models of volatility were identified in each city for each source market, as well as, different types of persistence of volatility, in each market and city, and different magnitude in face of good news and bad news, which strengthens the need to adjust the modelling of tourism demand for each market and, within a country, at a more detailed territorial scale.
The use of volatility models is quite recent in tourism demand modelling and had not yet been applied in cities in Portugal, for which, despite the growing importance in terms of tourism, there are no studies of modelling focusing on the tourism demand.
Modelling tourism demand is essential when tourism policymakers plan tourism activities. The tourism industry may be extremely sensitive to specific events’ effects, so good models must be found that reflect volatility that varies within each city and for each source market and policies must be adapted to each of the source/destination pairs.
... Using economic climate tracer, Andraz and Rodrigues (2016) show that the relationship between economic cycles and tourism flows is very small for Portugal over the period 1987-2015. Moreover, to examine trends and seasonality in tourism demand, researchers have widely used the univariate decomposition frameworks (Chan & Lim, 2011;Hassani et al., 2015;Kuo et al., 2008;Li & Law, 2020), while multivariate decomposition has mostly been neglected. Using the multivariate trend-cycle decomposition method, Vatsa (2020) examines the co-movement amongst the demand for New Zealand tourism. ...
This study examines the time–frequency causal effect of the novel coronavirus (COVID-19) pandemic on the European tourism sector. To do that, we use the wavelet coherence and phase difference framework, which enables us to investigate the causal interplays and business cyclical co-movement between COVID-19 crisis and the European tourism sector at several investment horizons. Our main results are as follows. First, the COVID-19 outbreaks significantly and highly affected the European tourism sector. Second, mid- and long-term causal effects of the recent outbreaks are greater than the effects on the short-term horizon. Third, the transmission channel from COVID-19 crisis to the European tourism sector decreases slowly as wavelet scales increase. Further, our empirical findings afford various policy implications to investors and tourism actors.
... Using economic climate tracer, Andraz and Rodrigues (2016) show that the relationship between economic cycles and tourism flows is very small for Portugal over the period 1987-2015. Moreover, to examine trends and seasonality in tourism demand, researchers have widely used the univariate decomposition frameworks (Chan & Lim, 2011;Hassani et al., 2015;Kuo et al., 2008;Li & Law, 2020), while multivariate decomposition has mostly been neglected. Using the multivariate trend-cycle decomposition method, Vatsa (2020) examines the co-movement amongst the demand for New Zealand tourism. ...
... Finally, RECI, which refers to structural breaks in time series data, may affect the forecasting process as unexpected environmental factors. The study by [33] supported the hypothesis that the recessions caused a delayed impact on tourist arrivals. Therefore, the above-explained inputs are used in this study to forecast tourist arrivals accurately. ...
This paper proposes a dynamic ensemble algorithm to combine forecasting results from multiple methodologies subject to their local (recent) predictive performance. In contrast to conventional combination forecasts, the proposed algorithm runs a sparsification process to merge a subset of methodology space to avoid overfitting and improve out-of-sample accuracy. The methodology space consists of various linear and non-linear as well as univariate and multivariate forecasting algorithms frequently used in the literature and industrial practice. The proposed algorithm continuously searches for the best combination to learn models weight. The weights are then used to combine the next forecasting coming from all forecasters. Two empirical studies are presented for illustrating its mechanism and predictive performance: crude oil price forecasting problem and tourist arrival forecasting. Empirical results support the fact that there is no one-fits-all methodology that outperforms in all periods. Our combination algorithm picks a different subset in each step, so the combination structure is dynamically redefined. Although some methodologies perform poorly, and they are never selected for the subset (e.g., ARIMA, ETS), most other methodologies are interchangeably picked or discarded from the combination structure.
... The ARFIMA (autoregressive fractional integrated moving average) model (Chu, 2009&Chu,2011, the ARIMA-GARCH (generalized autoregressive conditional heteroskedastic) model (Chan et al., 2005;Li, Chen, Wang,Ming, 2018& Chu,2011, and the SARIMA-In model (Chan, Lim, & McAleer, 2005;Li et al., 2018) are both dependent on ARIMA-type models (Goh & Law, 2002). Some techniques that were historically used in other fields (such as spectrum analysis or wavelet analysis) have also appeared in the tourism demand literature in the last five decades as a result of increased cross-disciplinary interaction Balli, F., Shahzad, S. J. H., & Uddin , 2018;Hassani,Webster, Silva & Heravi, 2015;S. Li et al., 2018). ...
This research examines 145 key papers from 1979 to 2020 in order to gain a better sense of how tourism demand forecasting techniques have changed over time. The three types of forecasting models are econometric, time series,
and artificial intelligence (AI) models. Econometric and time series models that were already popular in 2005 maintained their popularity, and were increasingly used as benchmark models for forecasting performance
assessment and comparison with new models. In the last decade, AI models have advanced at an incredible rate, with hybrid AI models emerging as a new trend. In addition, some new developments in the three categories of
models, such as mixed frequency, spatial regression, and combination and hybrid models have been introduced. The main conclusions drawn from historical comparisons forecasting methods are that forecasting models have
become more diverse, that these models have been merged, and that forecasting accuracy has improved. Given the complexities of predicting tourism demand, there is no single approach that works well in all circumstances, and forecasting techniques are still evolving.
... When we face non-stationary and complex time series, these non-linear patterns might remain after applying these methods. To improve prediction, after these seasonal components cleaning methods, several decomposition techniques are widely used for forecasting, such as the exponential smoothing methods (ETS) (Bergmeir, Hyndman, and Benítez 2016), or singular spectrum analysis (SSA) models (Hassani et al. 2015). All these techniques try to decompose time series into trend, cycle and a remainder component. ...
Johansen’s Cointegration Test (JCT) performs remarkably well in finding stable bivariate cointegration relationships. Nonetheless, the JCT is not necessarily designed to detect such relationships in presence of non-linear patterns such as structural breaks or cycles that fall in the low frequency portion of the spectrum. Seasonal adjustment procedures might not detect such non-linear patterns, and thus, we expose the difficulty in identifying cointegrating relations under the traditional use of JCT. Within several Monte Carlo experiments, we show that wavelets can empower more the JCT framework than the traditional seasonal adjustment methodologies, allowing for identification of hidden cointegrating relationships. Moreover, we confirm these results using seasonally adjusted time series as US consumption and income, gross national product (GNP) and money supply M1 and GNP and M2.
... Chen and Wu [33] applied EMP on tourist arrival in Taiwan. Singular Spectral Analysis (SSA) was used by Hassani et al. [34] for US tourist arrival and claimed increased performance. Silva et al. [35] used SSA and other decomposition with the neural network to forecast tourism demand for 10 European countries, namely Germany, Greece, Spain, Italy, Cyprus, Netherland, Austria, Sweden, and the UK. ...
Tourism makes a significant contribution to the economy of almost every country, so accurate demand forecasting can help in better planning for the government and a range of stakeholders involved in the tourism industry and can aid economic sustainability. Machine learning models, and in particular, deep neural networks, can perform better than traditional forecasting models which depend mainly on past observations (e.g., past data) to forecast future tourist arrivals. However, search intensities indices (SII) indicators have recently been included as a forecasting model, which significantly enhances forecasting accuracy. In this study, we propose a bidirectional long short‐term memory (BiLSTM) neural network to forecast the arrival of tourists along with SII indicators. The proposed BiLSTM network can remember information from left to right and right to left, which further adds more context for forecasting in memory as compared to a simple long short term memory (LSTM) network that can remember information only from left to right. A seasonal and trend decomposition using the Loess (STL) approach is utilized to decompose time series tourist arrival data suggested by previous studies. The resultant approach, called STL‐BiLSTM, decomposes time series into trend, seasonality, and residual. The trend provides the general direction of the overall data. Seasonality is a regular and predictable pattern which re‐occurs at fixed time intervals, and residual is a random fluctuation that is something which cannot be forecast. The proposed BiLSTM network achieves better accuracy than the other methods considered under the current study.
... Yao [4] used a neural networkenhanced hidden Markov time series forecasting model to predict the number of inbound tourists in the United States. Hassani [5] used Singular Spectrum Analysis (SSA) to forecast the monthly travel demand in the United States. Models for tourist arrivals focus on long-term prediction (e.g monthly, quarterly, and annual) of relatively large areas (e.g provinces, countries, and regions). ...
To improve the forecasting accuracy of tourism demand through forecasting model and data sources, this paper takes the social network data as an entry point, and collects the social network data by the web crawler, then quantifies the data based on the sentiment analysis of the BERT model. This paper uses structured variables such as social network data, weather, holidays, etc. to build a tourism demand forecasting model based on Gradient Boosting Regression Trees. At last, take Huang Shan as example, use actual statistics of passenger terminal and social network data to do an empirical analysis of Huang Shan tourism demand forecasting. Compared with the existing model and introduce ablation study to verify the effectiveness of the considered factors. The result shows that the model based on social network data has improved the forecasting accuracy from the existing ones, ablation study shows social network data helps to improve the accuracy of tourism demand forecasting.
... Forecasting foreign tourist arrivals is very important for the purposes of planning, policy making, and budgeting by parties related to tourism [8]. Over the year's various time series models have been successfully applied to predict tourist arrivals [9]. To indicate the level of foreign tourist visits requires a forecast with the right method, namely the Time series model, which is a method of forecasting widely used in various fields. ...
... Furthermore, Hassani et al. examined the possible benefits of tourism demand forecasting utilizing SSA [11]. The authors evaluated SSA forecasting ability on United States tourist arrivals monthly data from 1996 to 2012. ...
The paper delivers an assessment of Singular Spectrum Analysis (SSA) forecasting ability for short- and medium-term forecasting horizon, on real time traffic volume data. The key study goal is to estimate forecasting pertinency for daily traffic volume, based upon measurements at toll station. The suggested methodology is tested on real data from Moschohorion and Pelasgia Toll Station – Greece, utilizing custom developed forecasting software toolbox. Applied research results confirm an advanced forecasting ability of proposed methodology for short-term forecasting horizon against medium term forecasting horizon, when performance is compared upon the statistical criteria of the coefficient of determination R2. The obtained results present that SSA forecasting model could provide a competent forecasting methodology for road traffic volume data.
... These methods mainly model the nonlinear nature of tourism demand. The research results show that when the data possess nonlinear characteristics, artificial intelligent methods can effectively improve the model prediction capabilities compared with the benchmark models (Hassani et al., 2015;Law and Au, 1999;Law et al., 2019;Pai and Hong, 2005;Palmer et al., 2006;Zhang et al., 2020). However, no fixed prediction technology has excelled in all circumstances (Song and Li, 2008). ...
The large amounts of hospitality and tourism-related search data sampled at different frequencies have long presented a challenge for hospitality and tourism demand forecasting. This study aims to evaluate the applicability of large panels of search series sampled at daily frequencies to improve the forecast precision of monthly hotel demand. In particular, a hybrid mixed-data sampling regression approach integrating a dynamic factor model and forecast combinations is the first reported method to incorporate mixed-frequency data while remaining parsimonious and flexible. A case study is undertaken by investigating Sanya, the southernmost city in Hainan province, as a tourist destination using 9 years of the experimental data set. Dynamic factor analysis is used to extract the information from large panels of web search series, and forecast combinations are attempted to obtain the final prediction results of the individual forecasts to enhance the prediction accuracy further. The empirical analysis results suggest that the developed hybrid forecast approach leads to improvements in monthly forecasts of hotel occupancy over its competitors.
... In recent years, many researchers have been exploring the different market impact factors driving the volatility of the tourism market at different time frequencies (Goh, 2012;Pizam & Fleischer, 2002;Wang, 2009;. Accordingly, several authors have paid more attention to the application of data decomposition methods for analyzing tourism demand from the perspectives of time frequencies, such as spectral analysis (Coshall, 2000;Koži c, 2014), singular spectrum analysis (SSA; Beneki et al., 2012;Hassani et al., 2015), Fourier decomposition (Apergis et al., 2017), wavelet decomposition (Kummong & Supratid, 2016) and the empirical mode decomposition (EMD) family (Chen et al., 2012;Li et al., 2016;Zhang, Wu, et al., 2017a). Table 1 summarizes the related literature that incorporated data decomposition methods to analyze and forecast tourism demand. ...
Previous studies have shown that different market factors influence tourism demand at different timescales. Accordingly, we propose the decomposition ensemble learning approach to analyze impact of different market factors on tourism demand, and explore the potential advantages of the proposed method on forecasting tourism demand in Asia‐Pacific region. By decomposing tourist arrivals with noise‐assisted multivariate empirical mode decomposition, this study further explores the multiscale relationship between tourist destinations and major source countries. The empirical results show that decomposition ensemble approach performs significantly better than benchmarks in terms of the level forecasting accuracy and directional forecasting accuracy.
Purpose
The aim of this paper is to provide a narrative review of previous research on tourism demand modelling and forecasting and potential future developments.
Design/methodology/approach
A narrative approach is taken in this review of the current body of knowledge.
Findings
Significant methodological advancements in tourism demand modelling and forecasting over the past two decades are identified.
Originality/value
The distinct characteristics of the various methods applied in the field are summarised and a research agenda for future investigations is proposed.
Introduction Singular spectrum analysis (SSA) is a new powerful method in time series analysis. This non-parametric method is very suitable due to its unique features, such as not requiring the assumptions of time series stationarity, normality of residuals, and its applicability to short time series. The main purpose of SSA method is to decompose time series into interpretable components such as trends, oscillatory components, and unstructured noise. The basis of SSA is the singular value decomposition of the trajectory matrix built on the time series. In the basic SSA method the frequency of observations which used in the trajectory matrix is different and so there may be an error in reconstructing and forecasting the time series, especially at the beginning and end of the series. It occurs because the magnitude of eigenvalues, eigenvectors, and consequently, reconstruction and forecasting of future values of time series, is directly related to the trajectory matrix. The purpose of this paper is to improve the trajectory matrix of the SSA method to increase the accuracy of the reconstructed time series and forecasting results, which is called singular spectrum decomposition (SSD). In this paper, SSA and SSD methods and their properties are briefly introduced and then the performance of SSD method over SSA method in time series reconstruction and forecasting for simulated and real data is discussed.
Rainfall is a natural phenomenon that depends on many factors that are an important part of life on earth. The high intensity of rainfall can lead to disasters. Therefore, this study aims to forecast monthly rainfall. The data used was obtained from BMKG Bali Province, namely monthly rainfall data for Denpasar City from 2009 to 2019. The method used is backpropagation artificial neural network. The artificial neural network method is an information processing method inspired by the human nervous system. Optimal backpropagation network architecture is needed so that the prediction results have a low error rate, by optimizing the use of training data and test data taken from sample data. Based on the results of the testing and prediction process with the parameters of one hidden layer with 50 neorons, epoch 11 and learning rate 0.01, the results obtained with the MSE value in network testing are 0.037. So it can be concluded that the backpropagation artificial neural network method has good accuracy results used as a reference for decision making in predicting monthly rainfall in Denpasar City in the future.
In the paper, we propose a flexible AutoML approach that combines multiple feature generation strategies (spectral, wavelet, topological, quantile) and classifiers as parts of the modelling pipeline. It allows obtaining a more robust and lightweight solution for the time series classification problem. Generation of the pipeline is based on an evolutionary algorithm.
Compared with coarse-grained forecasting, fine-grained tourism demand forecasting is a more challenging task, but research on this issue is very scarce. To address this issue, a decomposition ensemble deep learning model is proposed by integrating CEEMDAN, CNNs, LSTM networks, and AR models. The CEEMDAN can decompose complex tourism demand data into multiple components with simpler characteristics, thereby reducing the complexity of forecasting. The CNNs and LSTM networks can fully capture the locally recurring patterns and the long-term dependencies of the components obtained by CEEMDAN. The AR model can capture the scale of tourism demand data, which can overcome the problem that the output scale of the deep neural networks (i.e., CNNs and LSTM networks) is not sensitive to the scale of the inputs. The effectiveness of the proposed model is verified by comparing with five benchmark models using real-time data on tourist volumes at two attractions.
Salinity is a classic problem in water quality management since it is directly associated with low water quality indices. Debate continues about selecting the best model for water quality forecasting, it remains a major challenge and causes much uncertainty. Accordingly, identifying the optimal modelling that can capture the salinity behavior is becoming a common trend in recent water quality research. This study applies novel combined techniques, including data preprocessing and artificial neural network (ANN) optimized with constriction coefficient-based particle swarm optimization and chaotic gravitational search algorithm (CPSOCGSA) to forecast monthly salinity data. Historical monthly total dissolved solids (TDS) and electrical conductivity (EC) data of the Euphrates River at AlMusayyab, Babylon, and climatic factors from 2010 to 2019 were used to build and validate the methodology. Additionally, for more validation, the CPSOCGSA-ANN was compared with the slime mould algorithm (SMA-ANN), particle swarm optimization (PSO-ANN) and multi-verse optimizer (MVO-ANN). The results reveal that the preprocessing data approaches improved data quality and selected the best predictors’ scenario. The CPSOCGSA-ANN algorithm is the best based on several statistical criteria. The proposed methodology accurately simulated the TDS and EC time series based on R2 = 0.99 and 0.97, respectively, and SI = 0.003 for both parameters.
This paper evaluates the impact of high-speed rail systems on tourist attractiveness in Italy. The analysis is carried out with reference to provincial capitals, only some of which are served by high-speed railway lines. To achieve this objective, two multiple linear regression models were specified and calibrated, which relate arrivals and presences in accommodation facilities to several factors that could influence the tourist destination: cultural, historical, and monumental heritage, commercial activities, recreational activities, accessibility, etc. Both models showed that the availability of high-speed railway services is an important factor in the choice of tourist destination, being, moreover, the only accessibility variable found to be significant; furthermore, the elasticity of tourist demand to this factor was significant too.
Water level (WL) forecasting has become a difficult undertaking due to spatiotemporal fluctuations in climatic factors and complex physical processes. This paper proposes a novel hybrid machine learning model based on an artificial neural network (ANN) and the Marine Predators algorithm (MPA) for modeling monthly water levels of the Tigris River in Al-Kut, Iraq. Data preprocessing techniques are employed to enhance data quality and determine the optimal input model. Historical data for water level and climatic factors data are utilized from 2011 to 2020 to build and assess the model. MPA-ANN algorithm’s performance is compared with recent constriction coefficient-based particle swarm optimization and chaotic gravitational search algorithm (CPSOCGSA-ANN) and slime mold algorithm (SMA-ANN) to reduce uncertainty and raise the prediction range. The finding demonstrated that singular spectrum analysis is a highly effective method to denoise time series. MPA-ANN outperformed CPSOCGSA-ANN and SMA-ANN algorithms based on different statistical criteria. The suggested novel methodology offers good results with scatter index (SI) = 0.0009 and coefficient of determination (R2 = 0.98).
Previous researches have proposed five strategies to deal with complex multi-step ahead forecasting tasks. However, these strategies have not received much attention in the field of tourism research and the performance of them is still unknown. Accordingly, we summarize the strategies used in multi-step ahead tourism demand forecasting articles and produce a comparative analysis of five strategies. By employing nine tourist arrival time series in Hong Kong, the study explores the performance of different strategy-driven forecasting approaches in tourism demand. The empirical results show that Direct and DIRMO (s=2) strategies perform better than other strategies in terms of forecasting accuracy, but there is no significant difference between them, excepting the latter is lower than the former in computational complexity.
We introduce two forecasting methods based on a general class of non-linear models called ‘State-Dependent Models’ (SDMs) for tourism demand forecasting. Using a Monte Carlo simulation which generated data from linear and non-linear models, we evidence how estimations from SDMs can capture the level shifts pattern and nonlinearity in data. Next, we apply two new forecasting methods based on SDMs to forecast tourism demand growth in Japan. The forecasts are compared with classical recursive SDM forecasting, Naïve forecasting, ARIMA, Exponential Smoothing, Neural Network models, Time varying parameters, Smooth Transition Autoregressive models, and with a linear regression model with two dummy variables. We find that improvements in forecasting with the proposed SDM-based forecasting methods are more pronounced in the longer-term horizons.
Effective and timely forecasting of daily tourism volume is an important topic for tourism practitioners and researchers, which can reduce waste and promote the sustainable development of tourism. Several studies are based on the decomposition-ensemble model to forecast the time series of high volatility in tourism volume, but ignore different forecasting methods suitable for different subseries. This study provides an adaptive decomposition-ensemble hybrid forecasting approach. Firstly, complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is used to effectively decompose the original time series into multiple relatively easy subseries, which reduces the complexity of the data. Secondly, sample entropy calculates the complexity of a sequence, and then adopts the elbow rule to adaptively divide them into different complex sets. Finally, multi-kernel extreme learning machine (KELM) models are used to forecast the components of different sets and integrate them. This hybrid approach makes full use of the advantages of different models, which enables effective use of data. The empirical results demonstrate that the approach can both produce results that are close to the actual values and be utilized as a strategy for forecasting daily tourism volume.
This study examines the time–frequency causal effect of the novel coronavirus (COVID-19) pandemic on the European tourism sector. To do that, we use the wavelet coherence and phase difference framework, which enables us to investigate the causal interplays and business cyclical co-movement between COVID-19 crisis and the European tourism sector at several investment horizons. Our main results are as follows. First, the COVID-19 outbreaks significantly and highly affected the European tourism sector. Second, mid- and long-term causal effects of the recent outbreaks are greater than the effects on the short-term horizon. Third, the transmission channel from COVID-19 crisis to the European tourism sector decreases slowly as wavelet scales increase. Further, our empirical findings afford various policy implications to investors and tourism actors.
The tourism sector has an important role in helping the income of a region, especially for economic development and opportunities to expand employment. However, the trend tourist arrival to these tourist attractions has decreased since the COVID-19 pandemic. The government enforces a new normal policy to reopen tourist attractions by implementing health protocols. Local governments and tourism managers need forecasting of tourist arrivals to help plan the tourism sector in the future and anticipate an increase in tourist arrival. Most tourist arrivals are influenced by several factors, such as : seasonality, politics, disasters, crises, and other important events. One method to accommodate these factors is using Ensemble Empirical Mode Decomposition (EEMD). However, EEMD still produces a mixing mode during decomposition. Complete Ensemble Mode Decomposition with Adaptive Noise (CEEMDAN) is proposed to overcome the weaknesses of EEMD. This research integrates CEEMDAN with Feedforward Neural Network (FNN) in generating forecasts. The experiment results show that the integration of CEEMDAN and FNN can produce good forecasting accuracy.
In this study, we propose an adaptive multiscale ensemble (AME) learning approach, which consists of variational mode decomposition (VMD) and least square support vector regression (LSSVR) for seasonal and trend forecasting of tourist arrivals. In the formulation of AME learning approach, the original tourist arrival series is decomposed into the trend, seasonal, and remaining volatility components. Then, ARIMA is used to forecast the trend component, SARIMA is used to forecast the seasonal component, and LSSVR is used to forecast the remaining volatility components. The empirical results demonstrate that our proposed AME learning approach can achieve higher forecasting accuracy.
Abstrak-Pendidikan Sekolah Dasar (SD) sangat penting dalam memberikan keterampilan dasar yang dibutuhkan siswa untuk bertahan dalam mengikuti dan memahami kelas-kelas pada jenjang di atasnya sehingga jika pondasi pendidikan SD kuat maka dapat meningkatkan kualitas pendidikan di Indonesia. Tujuan dari penelitian adalah membuat model arsitektur terbaik yang akan digunakan untuk melakukan peramalan pada jumlah siswa mengulang berdasarkan provinsi pada jenjnag Sekolah Dasar baik negeri dan swasta dengan model berbasis sistem kecerdasan buatan. Sumber data berasal dari data statistik pendidikan dengan url: http://statistik.data.kemdikbud.go.id/. Data terdiri dari 34 provinsi untuk tahun ajaran. Metode solusi yang digunakan adalah back-propagation yang merupakan bagian dari sistem kecerdasan buatan dimana dalam menentukan model arsitektur terbaik dilakukan dengan menguji serangkaian arsitektur (2-5-1; 2-10-1; 2-15-1 dan 2-20-1) mengunakan fungsi aktivasi sigmoid dan parameter pendukung seperti performFcn = MSE; goal = 0.001; epochs = 10000; mc = 0.95 dan lr = 0.1. Hasil menunjukkan back-propagation dapat diterapkan untuk melakukan peramalan dengan sistem kecerdasan buatan dengan menghasilkan sebuah model arsitektur terbaik yakni 2-10-1 dengan MSE pelatihan adalah 0.00099299, koefisien korelasi (R) pelatihan adalah 0.976972 pada epoch 81, MSE pengujian adalah 0.001325, koefisien korelasi (R) pengujian dengan akurasi 85%. Fakta baru menyebutkan bahwa akurasi sangat tergantung terhadap banyaknya data. Selain harus memperhatikan MSE, epoch dan durasi waktu pelatihan. Diharapkan model arsitektur ini dapat membantu melakukan peramalan terhadap jumlah siswa mengulang pada jenjang Sekolah Dasar sehingga hasil dari peramalan dapat digunakan untuk membuat analisis kebijakan.
Tourism trend prediction facilitates estimation of tourism investment and revenue. Studies on tourism prediction have primarily relied on linear models and historical visitors; however, relationships between tourism trends and their factors may be nonlinear. This study constructed factors from internet search data and predicted tourism trends using a spatiotemporal framework based on the extreme gradient boosting (XGBoost) method. The study first sorted Baidu index data that is computed by weighting the search frequency. The spatial cluster analysis was conducted to incorporate spatial characteristics, and principal component analysis was further performed to identify factors. The next step derived variables using the weighted moving average method to reduce the lag effect between tourism internet search and actual behavior. We applied the proposed spatiotemporal XGBoost composite model to predict Beijing’s tourism trends. The R² scores of the simple XGBoost model, the autoregressive integrated moving average model, the spatial XGBoost model, and the spatiotemporal XGBoost composite model were 0.517, 0.625, 0.791, and 0.940, respectively. Compared to predictions from different models, the spatiotemporal XGBoost composite model has the best prediction ability. The findings also suggest that machine learning methods may not perform well without considering spatial properties, such as spatial autocorrelation and spatial heterogeneity.
The demand for international tourism in Australia and New Zealand is vital to the South Pacific’s tourism-reliant islands. However, at the time of this study these two countries find themselves in precarious economic situations. The question addressed by this study is, will tourism imports in these two countries pick up on the back of economic recovery? We answer this question using time-difference analysis and the newly developed Hamilton filter. The short answer is yes, but more so in New Zealand than Australia. The key findings of this study are that tourism demand in both Australia and New Zealand is pro-cyclical, tourism demand cycles in New Zealand strongly lag business cycles by 1 year, whereas in Australia, they weakly lag business cycles by one quarter, and, overall, tourism demand and business cycles in New Zealand share a stronger association than they do in Australia.
This study aimed to determine the accuracy of forecasting methods for trend and seasonal Indonesia visitor arrival. Three single methods, Seasonal Autoregressive Moving Average (SARIMA), Singular Spectrum Analysis (SSA), and Fuzzy Time Series (FTS) were used to model and predict the monthly arrival. We have also compared the three single methods with two hybrid approaches, SARIMA-FTS, and SSA-FTS. Results show that SARIMA-FTS is the most appropriate model to capture the trend and seasonal pattern of the series in term of Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE).
This article builds on the work of Greenidge, employing structural time series model (STSM) to explain and forecast quarterly tourist flows from Barbados' primary source markets — the USA, the United Kingdom, Canada and CARICOM — for the period 1966:1-2007:4. Results show that the structure and nature of the individual time series components have evolved significantly since the work of Greenidge. Of particular interest, arrivals from the main source markets appear to be less income sensitive. The study also investigates the predictive power of STSM. A seasonal naive model is used for benchmark comparison purposes. We find that STSM outperforms the seasonal naive model in its both multivariate and univariate form.
In recent years Singular Spectrum Analysis (SSA), a relatively novel but powerful technique in time series analysis, has been developed and applied to many practical problems across different fields. In this paper we review recent developments in the theoretical and methodological aspects of SSA from the perspective of analyzing and forecasting economic and financial time series, and also present some new results. In particular, we (a) show what are the implications of SSA for the frequently invoked unit root hypothesis of economic and financial times series; (b) introduce two new versions of SSA, based on the minimum variance estimator and based on perturbation theory; (c) discuss the concept of causality in the context of SSA; and (d) provide a variety of simulation results and real world applications, along with comparisons with other existing methodologies.
The different forms of the multivariate singular spectrum analysis (SSA) and their associated forecasting algorithms are considered from both theoretical and practical points of view. The new multivariate vector forecasting algorithm is introduced and its uniqueness is evaluated. The performance of the new multivariate forecasting algorithm is assessed against the existent multivariate technique using various simulated and real data sets (namely European Electricity and Gas series). The forecasting results confirm that the performance of the new multivariate approach is more accurate than the current approach. The optimality of the window length and the number of eigenvalues in multivariate SSA are considered and various bounds are recommended. The effect of common components between two time series is evaluated through a simulation study. The concept of similarity and dissimilarity are also considered based on the matched components among series.
This paper investigates the impact of the Great Japan Earthquake (and subsequent tsunami turmoil) on socio-economic activities by using data on hotel opportunities collected from an electronic hotel booking service. A method to estimate both primary and secondary regional effects of a natural disaster on human behavior is proposed. It is confirmed that temporal variation in the regional share of available hotels before and after a natural disaster may be an indicator to measure the socio-economic impact at each district.
In this article, we propose a new set of multivariate stochastic models that capture time-varying seasonality within the vector innovations structural time-series (VISTS) framework. These models encapsulate exponential smoothing methods in a multivariate setting. The models considered are the local level, local trend, and damped trend VISTS models with an additive multivariate seasonal component. We evaluate the forecasting accuracy of these models against the forecasting accuracy of univariate alternatives using international tourist arrivals from 11 source countries to Australia and New Zealand. In general, the newly proposed multivariate models improve on forecast accuracy over the univariate alternatives.
The recent economic crisis and swine flu pandemic have had significantly negative impacts on global tourism. Tourism in the United Kingdom has also suffered as a result of the two crises, although their actual impacts have yet to be evaluated. This study analyzes the impacts of these two phenomena on the demand for U.K. inbound tourism during the 2008Q1-2009Q2 period among visitors from the country’s 14 major visitor source markets. An econometric framework is proposed to separate and estimate the impacts of swine flu and the economic crisis on U.K. tourism demand.
This paper aims to provide the most up-to-date survey of tourism economics research and to summarise the key trends in its recent development. Particular attention is paid to the research progress made over the last decade in respect of approaches, methodological innovations, emerging topics, research gaps, and directions for future research. Remarkable but unbalanced developments have been observed across different sub-research areas in tourism economics. While neoclassical economics has contributed the most to the development of tourism economics, alternative schools of thought in economics have also emerged in advancing our understanding of tourism from different perspectives. As tourism studies are multi- and inter-disciplinary, integrating economics with other social science disciplines will further contribute to knowledge creation in tourism studies.
The incessantly growing demand for energy consumption and the signi�cance of the
availability of sustainable energy for achieving long term economic growth de�nes the
importance of forecasting energy statistics. This paper analyses and forecasts actual
energy consumption data for EU-27 nations using both parametric and nonparametric
time series forecasting techniques. Singular Spectrum Analysis (SSA) is adopted as the
nonparametric time series analysis and forecasting technique and the results from SSA
are compared with ARIMA, which is a parametric forecasting technique.
Keywords: Electricity consumption; Renewable electricity consumption; Primary energy
consumption; Forecasting; Singular Spectrum Analysis; ARIMA; Parametric; Nonpara-
metric.
This study investigates the performance of combination forecasts in the context of international tourism demand. Five econometric and two time-series models are employed to generate individual forecasts. Six combination methods are then employed, and their forecasting performance evaluated, using data on UK outbound tourism demand in seven destination countries. The results suggest that combination forecasts, in general, outperform the best individual forecasts. More sophisticated such combination methods, which take the historical performance of individual forecasts into account, perform better than the simple average technique. The performance of combination forecasts is associated with the performance consistency of the individual forecasts they include, and the inclusion of up to three individual forecasts is most likely to result in accurate combination forecasts.
This paper reviews the published studies on tourism demand modelling and forecasting since 2000. One of the key findings of this review is that the methods used in analysing and forecasting the demand for tourism have been more diverse than those identified by other review articles. In addition to the most popular time-series and econometric models, a number of new techniques have emerged in the literature. However, as far as the forecasting accuracy is concerned, the study shows that there is no single model that consistently outperforms other models in all situations. Furthermore, this study identifies some new research directions, which include improving the forecasting accuracy through forecast combination; integrating both qualitative and quantitative forecasting approaches, tourism cycles and seasonality analysis, events' impact assessment and risk forecasting. r 2007 Elsevier Ltd. All rights reserved.
Accurate forecasts of tourist arrivals and study of the tourist arrival patterns are essential for the tourism-related industries to formulate efficient and effective strategies on maintaining and boosting tourism industry in a country. Forecasting accuracy is one of the most important factors involved in selecting a forecasting method. This study presents a hybrid artificial intelligence (AI) model to develop a Mamdani-type fuzzy rule-based system to forecast tourist arrivals with high accuracy. The hybrid model uses genetic algorithm for learning rule base and tuning data base of fuzzy system. Actually it extracts useful information patterns with a descriptive rule induction approach based on Genetic Fuzzy Systems (GFS). This is the first study on using a GFS with the ability of learning rule base and tuning data base of fuzzy system for tourist arrival forecasting problem. Evaluation of the proposed approach will be carried out by applying it to a case study of tourist arrivals to Taiwan and results will be compared with other studies which have used the same data set. Results show that the proposed approach has high accuracy, so it can be considered as a suitable tool for tourism arrival forecasting problems.
Africa's tourism potential is acknowledged to be significant but underdeveloped. This paper uses both cross-section data and panel data for the period 1996-2000 to identify the determinants of tourism arrivals in 43 African countries, taking into account tourists' country of origin. The results strongly suggest that political stability, tourism infrastructure, marketing and information, and the level of development at the destination are key determinants of travel to Africa. Typical 'developed country determinants' of tourism demand, such as the level of income in the origin country, the relative prices and the cost of travel, are not so significant in explaining the demand for Africa as a tourism destination. It is therefore recommended that attention should be given to improving the overall stability of the continent and the availability and quantity of tourism infrastructure.
Empirical evidence has shown that seasonal patterns of tourism demand and the effects of various influencing factors on this demand tend to change over time. To forecast future tourism demand accurately requires appropriate modelling of these changes. Based on the structural time series model (STSM) and the time-varying parameter (TVP) regression approach, this study develops the causal STSM further by introducing TVP estimation of the explanatory variable coefficients, and therefore combines the merits of the STSM and TVP models. This new model, the TVP-STSM, is employed for modelling and forecasting quarterly tourist arrivals to Hong Kong from four key source markets: China, South Korea, the UK and the USA. The empirical results show that the TVP-STSM outperforms all seven competitors, including the basic and causal STSMs and the TVP model for one- to four-quarter-ahead ex post forecasts and one-quarter-ahead ex ante forecasts.
This paper seeks to analyse and evaluate Sri Lanka's energy consumption forecasts, more specifically, electricity, petroleum, coal and renewable electricity consumption using a variety of parametric and nonparametric forecasting techniques. The Sri Lankan economy is emerging following the end of a prolonged civil war, and thus this topic is opportune as accurate forecasts of energy requirements are indispensable for sustaining the ongoing rapid economic expansion. We also consider evaluating the appropriateness of parametric and nonparametric energy forecasting methods for Sri Lanka. In addition, this paper marks the introductory application of SSA for forecasting in Sri Lanka, and the results from SSA are compared against the popular benchmarks of ARIMA, ETS, HW, TBATS, and NN. We find statistically significant evidence proving that the SSA model outperforms ETS and HW at forecasting renewable electricity consumption, and also, that the SSA model outperforms ARIMA and TBATS models at forecasting coal consumption. On average, the SSA model is found to be best for energy consumption forecasting in Sri Lanka whilst the Neural Networks model is second best.
This paper seeks to provide an alternative forecast to that provided by the Energy Information Administration (EIA) on energy-related monthly CO2 emissions in the United States. The data on CO2 emissions from petroleum, natural gas, coal and total fossil fuels obtained via the EIA covering the period January 2005 to November 2013 is analysed and then forecasted using ARIMA, Holt-Winters, and Exponential Smoothing prior to introducing the Singular Spectrum Analysis (SSA) technique for CO2 emissions forecasting. A new combination forecast (EIA-SSA) is also introduced by merging the SSA and EIA forecasts, and is seen outperforming all models including the EIA forecast. Finally, the EIA-SSA model is used to provide an alternative 12 month ahead outlook for US energy-related CO2 emissions from December 2013 to November 2014. This research is expected to influence the methodology adopted by the EIA for forecasting CO2 emissions in the future by improving the accuracy of the forecasts, and the impact of this study will be clearer upon comparing the actual CO2 emissions in US with the EIA, and EIA-SSA forecasts over the 12 month period which follows.
The aim of this article is to use dynamic panel data cointegration technique to determine elasticities of tourist arrivals to Australia, using income, real exchange rates, and airfares as demand determinants. Annual data from 1991 to 2007 for arrivals from the 10 main markets are used. Previous studies that applied dynamic panel data sets in the tourism context have used the Arellano—Bond estimation technique. Because this technique produces biased and inconsistent estimates in samples with a small time span, this article uses the corrected least square dummy variable technique to generate unbiased and efficient parameter estimates. The results obtained show that demand is inelastic with respect to its determinants in the short run and elastic in the long run. The main implications of these results are that maintaining destination price competitiveness and consumer satisfaction should be rated very high in the priorities of the Australian tourism industry.
This study introduces the concepts and theories relating to conditional heteroscedastic volatility models and the news impact curve, and applies them in the analysis of the tourist market in Korea. Three volatility models are used to estimate the conditional volatility of monthly arrivals of inbound tourists into Korea, and news impact curves are examined in the context of these models. The major findings of this study include the existence of monthly seasonality in conditional mean equations, the existence of asymmetric effects from the EGARCH and TARCH models, and the persistence of the impact of news shock on monthly tourist arrivals into Korea in the estimation of the GARCH model.
This paper generates short-term forecasts on tourist arrivals in Greece and performs impulse response analysis to measure the impact of macroeconomic shocks from the origin country on future tourism demand. We find the ARIMA (1, 1, 1) model outperforms exponential smoothing models in forecasting the direction of one year out of sample forecasts. However, this does not translate into point forecasting accuracy. Impulse response analysis on the impact of unemployment and tourists’ cost of living shocks shows that the source of downside risk to future tourism numbers is limited in scope, magnitude, and duration. Shocks to consumer confidence from the origin countries have no impact on future tourism demand. Our results offer important insights and implications for policymakers and tourist operators.
In this paper, we construct and use a piecewise linear method to model and forecast, on a monthly basis, the demand for Macau tourism. Data over the period January 1991–December 2005 and a seasonally adjusted series for tourism demand are used. The study examines 4 forecasting horizons ranging from 6 to 24 months in advance. Mean absolute percentage errors and root mean square errors are adopted as criteria for evaluating the accuracy of the forecasting exercises. Finally, the forecasts of piecewise linear model are compared with those of autoregressive trend model, seasonal autoregressive integrated moving average and its arch-rival fractionally integrated autoregressive moving average models. The piecewise linear model is more accurate than the three benchmark models tested and the improvement is practically significant.
In the face of the world recession, this article concentrates on analyzing and forecasting demand for outbound traveling in Australia, Canada, United States, Japan, and the EU-15 countries in terms of tourism imports at constant prices and exchange rates. Drawing on an analysis of the macroeconomic effects of the global recession and its impact on tourism demand for outbound traveling, the article develops forecasts for the period from 2009 to 2010. For 2010, two scenarios are created to project demand for foreign travel. The study is rounded off by a discussion of forecast uncertainties and open questions.
Among the major overseas tourism markets for Singapore (besides the neighbouring ASEAN countries), Japan is the second largest. The number of Japanese visitors to Singapore had increased from 0.38 million in 1985 to 0.84 million in 1998, an increase of almost 121 per cent. In terms of the share of Japanese tourists in the domestic tourism market, it had only increased from 12.4 to 13.5 per cent. To attract more Japanese visitors, it helps to have a clear understanding of the pattern of Japanese tourist arrivals so that different promotion strategies can be devised and implemented. In this paper, we analyzed the pattern of arrivals by focusing on seasonal indices. The study found that the seasonal indices of Japanese tourist arrivals are stable over time. Policy implications of the finding were also discussed.
In a recent paper, Bai and Perron (1998) considered theoretical issues related to the limiting distribution of estimators and test statistics in the linear model with multiple structural changes. In this companion paper, we consider practical issues for the empirical applications of the procedures. We first address the problem of estimation of the break dates and present an efficient algorithm to obtain global minimizers of the sum of squared residuals. This algorithm is based on the principle of dynamic programming and requires at most least-squares operations of order OOT 2 for any number of breaks. Our method can be applied to both pure and partial structural change models. Second, we consider the problem of forming confidence intervals for the break dates under various hypotheses about the structure of the data and the errors across segments. Third, we address the issue of testing for structural changes under very general conditions on the data and the errors. Fourth, we address the issue of estimating the number of breaks. Finally, a few empirical applications are presented to illustrate the usefulness of the procedures. All methods discussed are implemented in a GAUSS program.
In recent years the singular spectrum analysis (SSA) technique has been further developed and applied to many practical problems. The aim of this research is to extend and apply the SSA method, using the UK Industrial Production series. The performance of the SSA and multivariate SSA (MSSA) techniques was assessed by applying it to eight series measuring the monthly seasonally unadjusted industrial production for the main sectors of the UK economy. The results are compared with those obtained using the autoregressive integrated moving average and vector autoregressive models.
We also develop the concept of causal relationship between two time series based on the SSA techniques. We introduce several criteria which characterize this causality. The criteria and tests are based on the forecasting accuracy and predictability of the direction of change. The proposed tests are then applied and examined using the UK industrial production series. Copyright
We use univariate and multivariate singular spectrum analyses to predict the inflation rate as well as changes in the direction of inflation time series for the United States. We use consumer price indices and real-time chain-weighted GDP price index series in these prediction exercises. Moreover, we compare our out-of-sample, $h$-step-ahead moving prediction results with other prediction results based on methods such as activity-based NAIRU Phillips curve, $AR(p)$, the dynamic factors model, and random walk models with the latter as a naive forecasting method. We use short-run (quarterly) and long-run (one to six years) time windows for predictions and find that multivariate singular spectrum analysis outperforms all other competing prediction methods. Also, we confirm the results of earlier studies that prediction of the inflation rate in the United States during the period of the ``Great Moderation" is less challenging compared to the more volatile inflationary period of 1970-1985.
We present and apply singular spectrum analysis (SSA), a relatively new, non-parametric and data-driven method for signal extraction (trends, seasonal and business cycle components) and forecasting of UK tourism income. Our results show that SSA slightly outperforms SARIMA and time-varying-parameter state space models in terms of root mean square error, mean absolute error and mean absolute percentage error forecasting criteria. Copyright © 2011 John Wiley & Sons, Ltd.
It has been demonstrated in the tourism literature that a combination of individual tourism forecasting models can provide better performance than individual forecasting models. However, the linear combination uses only inputs that have a linear correlation to the actual outputs. This paper proposes a non-linear combination method using multilayer perceptron neural networks (MLPNN), which can map the non-linear relationship between inputs and outputs. UK inbound tourism quarterly arrivals data by purpose of visit are used for this case study. The empirical results show that the proposed non-linear MLPNN combination model is robust, powerful and can provide better performance at predicting arrivals than linear combination models.
This paper investigates the possibility of improving the predictive ability of a tourism demand model with meteorological explanatory variables. The authors use as a case study the monthly British tourism demand for the Balearic Islands (Spain). For this purpose, a transfer function model and causal artificial neural network are fitted. The results are compared with those obtained by non-causal methods: an ARIMA model and an autoregressive neural network. The results indicate that incorporating meteorological variables can increase predictive power, although the most accurate prediction is obtained using a non-causal model - specifically, an autoregressive neural network.
Estimating tourism demand has become a challenge among researchers, as identification of key determinants is important for policymakers at a time when tourism has become the world's largest industry. Using a theoretical framework based on the gravity model, this paper models inbound tourism demand for South Africa to estimate price and income sensitivities as well as the impact of other important factors that affect tourist flows, such as the location of markets and socio-political factors. Given the non-stationary but cointegrated nature of the panel data, panel cointegration estimation techniques are employed. The results show that tourists are sensitive to price changes in South Africa and also to tourism price changes in competing destinations. The level of development, tourism infrastructure, distance (or transportation costs), common border and language are also found to affect arrivals. The results also indicate the need to conduct estimation by regional groupings for a better understanding of different markets.
In 2003, the Severe Acute Respiratory Syndrome (SARS) ravaged many Asian countries. The outbreak of SARS caused a crisis in the tourism industry in many parts of Asia in that year. The purpose of this study is to examine and to compare the post-SARS recovery patterns of inbound arrivals from Japan, Hong Kong and USA in Taiwan. Taking the cusp catastrophe model as its foundation, this study proposes a well-grounded approach to understanding the nature of the recovery processes and to explaining the difference between the recovery patterns displayed by arrivals from Japan and those from Hong Kong and USA. Implications regarding tourism promotion policies are drawn from the analysis.
Forecast combination is a well-established and well-tested approach for improving the forecasting accuracy. One beneficial strategy is to use constituent forecasts that have diverse information. In this paper we consider the idea of diversity being accomplished by using different time aggregations. For example, we could create a yearly time series from a monthly time series and produce forecasts for both, then combine the forecasts. These forecasts would each be tracking the dynamics of different time scales, and would therefore add diverse types of information. A comparison of several forecast combination methods, performed in the context of this setup, shows that this is indeed a beneficial strategy and generally provides a forecasting performance that is better than the performances of the individual forecasts that are combined. As a case study, we consider the problem of forecasting monthly tourism numbers for inbound tourism to Egypt. Specifically, we consider 33 individual source countries, as well as the aggregate. The novel combination strategy also produces a generally improved forecasting accuracy.
This paper introduces a web-based tourism demand forecasting system (TDFS) that is designed to forecast the demand for Hong Kong tourism, as measured by tourist arrivals, total and sectoral tourist expenditures, and the demand for hotel rooms. The TDFS process comprises three stages–preliminary data analysis, the generation of quantitative forecasts and judgmental adjustments–which correspond to the three key system components: the data module, the quantitative forecasting module and the judgmental forecasting module, respectively. These stages (modules) interact with one another. This paper focuses on a recent case study that illustrates the functional ability of the TDFS as a support system, providing accurate forecasts of the demand for Hong Kong tourism. Specifically, the quantitative forecasts are generated by the autoregressive distributed lag model, then adjusted by a panel of experts comprising postgraduate students and academic staff. The results show that this combination of quantitative and judgmental forecasts improves the overall forecasting accuracy.
The Indian Ocean tsunami of 26 December 2004 made clear the devastating short term impact in lost lives, as well as the longer term impact in lost livelihoods and an uncertain future. Tourism was one of the more obvious non-human casualties of this tragic natural phenomenon. Tourism is the principal industry in the Maldives, accounting for more than 30% of GDP and more than 60% of foreign exchange earnings. As over 55% of government tax revenue arises from tourism-related taxes, and monthly government budget financing depends largely on international tourist arrivals, the monitoring of daily, weekly and monthly international tourist arrivals is essential for fiscal policy evaluation. Over 600,000 tourists visited the Maldives in 2004. This paper examines the time series properties of monthly international tourist arrivals to the Maldives from eight major tourist source countries, namely Italy, Germany, UK, Japan, France, Switzerland, Austria and the Netherlands, from 1 January 1994 to 31 December 2003. Monthly international tourist arrivals and the associated uncertainty are estimated for the eight principal tourist source countries. Univariate and multivariate time series models of conditional volatility (or uncertainty) are estimated and tested. The conditional correlations are estimated and examined to ascertain whether there is specialization, diversification or segmentation in the international tourism demand shocks from the major tourism source countries to the Maldives. The estimated static conditional correlations for monthly international tourist arrivals, as well as for the respective transformed series, were found to be significantly different from zero, but also relatively low. This indicates that the government of the Maldives and the major tour operators that organise tourist vacations have to emphasise their marketing efforts independently of each tourist source country.
A study of factors affecting the number of visitors to Hawaii during the period 1993–2007 prompted by an observed waning of the U. S. mainland to Hawaii visitor market was conducted. Both time series and cross section analyses revealed that Log Gross State Product, Log Chained Airfare and Log Distance to Orlando, Florida were the most important predictor variables. A mixed model which modeled Log Chained Gross State Product, Log Chained Airfare, two recessions plus the September 11, 2001 effect in addition to other fixed effects and random state effects was used. Cross section (spatial) airfare elasticities on an annual basis were high and growing over time, but those estimated from the time series analysis (temporal) were much lower. Elasticities derived from Gross State Product were moderately high and very stable over time. To counteract the distance effect, stopovers in existing mainland resort cities when en route to Hawaii and other promotions to develop a stronger presence of a Hawaii image were recommended.
Since tourism is the most important economic activity in many Caribbean countries, the forecast of tourism is the key to predicting overall macroeconomic performance in this region. This study utilized Structural Time Series Modeling to explain and forecast tourist arrivals to Barbados from its major generating markets. The procedure will allow one to extract the maximum amount of information contained in the series on tourist arrivals while at the same time to include any other information relevant to forecasting arrivals. This study found these models offered valuable insights into the stylized facts of tourism behavior and provided reliable out-of-sample forecasts.RésuméLa prévision de la demande de tourisme: une approche par SCS pour la Barbade. Étant donné que le tourisme est l'activité économique la plus importante dans beaucoup de pays des Caraı̈bes, la prévision du tourisme est la clé de la prévision de la performance macroéconomique globale de ces pays. Cette étude utilise la technique de la modélisation de série chronologique structurale (SCS) pour expliquer et prévoir l'arrivée des touristes à la Barbade en provenance des pays qui lui sont des fournisseurs majeurs en touristes. Cette technique permet d'extraire le maximum d'informations sur les arrivées des touristes et en même temps d'inclure toute autre information utile pour prévoir les arrivées. On a trouvé que ces modèles jettent de la lumière sur les faits stylisés du comportement du tourisme et fournissent des prévisions fiables hors-échantillon.
Volatility, exponential smoothing, regression and Naïve 2 models are considered singly and in combination in terms of forecasting demand for international tourism. These models generate accurate predictions of tourism flows, but their prime utility is when combined with other models. Usually, models are combined by means of purely statistical criteria. We show that goal programming (GP) offers an alternative, flexible approach to model combination. GP offers planners a practical solution to tourism forecasting problems, since the method is more adaptable than conventional minimisation of prediction error, by permitting practitioners to prioritise a series of management related goals. Forecasters can focus on longer- and short-term goals, minimising forecast under- and over-estimation and/or concentrate on prediction errors in tourism flows at various times of the year.
Univariate volatility models are applied to UK tourism demand to the country's most popular international destinations. Volatility is a concept borrowed from Finance. The fact that significant volatility models are found for ten of the twelve destinations examined shows that the volatility concept has relevance to tourism demand. Volatility models are able to quantify the impacts of positive and negative shocks on tourism demand. The impacts of negative shocks vary in magnitude and duration according to the destination involved and the nature of the shock. The forecasting capability of these models has never been assessed in the tourism field. They are shown to generate highly accurate forecasts, but become optimal when combined with forecasts obtained from exponential smoothing models. Two methods of combining individual forecasts are considered. Bias in individual volatility and smoothing models and in combinations of them is examined.
While a mega-sport event is scheduled at least once every year somewhere in the world, these events are rare occurrences for the host cities and countries. The benefits of such events seem lucrative; the very fact that many countries bid to host these events suggests that the benefits – be they tangible or intangible – more often than not outweigh the costs. Using a standard gravity model of bilateral tourism flows between 200 countries from 1995 to 2006, this paper measures a very direct benefit of such mega-events: the increase in tourist arrivals to the host country. In general, the results suggest that mega-sport events promote tourism but the gain varies depending on the type of mega-event, the participating countries and whether the event is held during the peak season or off-season.
This paper presents the use of time series SARIMA and MARIMA with interventions in forecasting tourism demand using ten arrival series for Hong Kong. Augmented Dickey–Fuller tests indicated that all the series were seasonal nonstationary. Significant interventions such as relaxation of the issuance of out-bound visitors visas, the Asian financial crisis, the handover, and the bird flu epidemic were all empirically identified with significant test results and expected signs. The forecasts obtained using models that capture stochastic nonstationary seasonality and interventions, SARIMA and MARIMA with intervention analysis, are compared with other eight time series models and were found to have the highest accuracy.