No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Investigating Causal Relations by Econometric Models and Cross-Spectral Methods
... Causality offers a structured approach to capture the underlying relationships within multivariate time-series (Runge et al., 2019). In multivariate time-series, the observed variables are governed by causal relationships that dictate how changes in one variable affect others (Granger, 1969;Cheng et al., 2024b). For instance, the increase in temperature directly causes the duration for which the air conditioner stays on to be longer. ...
... Causal discovery uncovers causal relationships among interdependent variables in multivariate time-series (Assaad et al., 2022;Cheng et al., 2024b). The Granger causality test (Granger, 1969) has been widely adopted to infer causal relationships by testing whether past values of one variable improve the prediction of another. Recently, deep learning approaches have been used for more robust causal discovery (Tank et al., 2021;Cheng et al., 2023;2024a). ...
... A multivariate time-series is defined as a collection of timeseries obtained from N variables {x i } N i=1 , where N > 1. According to the Granger causality (Granger, 1969), x i causes x j if the past values of x i affect the future values of x j . In this case, x i is the cause of x j while x j is the effect of x i . ...
Utilizing the complex inter-variable causal relationships within multivariate time-series provides a promising avenue toward more robust and reliable multivariate time-series anomaly detection (MTSAD) but remains an underexplored area of research. This paper proposes Causality-Aware contrastive learning for RObust multivariate Time-Series (CAROTS), a novel MTSAD pipeline that incorporates the notion of causality into contrastive learning. CAROTS employs two data augmentors to obtain causality-preserving and -disturbing samples that serve as a wide range of normal variations and synthetic anomalies, respectively. With causality-preserving and -disturbing samples as positives and negatives, CAROTS performs contrastive learning to train an encoder whose latent space separates normal and abnormal samples based on causality. Moreover, CAROTS introduces a similarity-filtered one-class contrastive loss that encourages the contrastive learning process to gradually incorporate more semantically diverse samples with common causal relationships. Extensive experiments on five real-world and two synthetic datasets validate that the integration of causal relationships endows CAROTS with improved MTSAD capabilities. The code is available at https://github.com/kimanki/CAROTS.
... The challenges and importance of causal discovery are amplified when analyzing time-series data. Temporal data introduces complexities such as lagged effects (where a cause at time − influences an effect at time ), feedback loops (where variables influence each other over time), and dynamic interactions that evolve [23]. For example, in climate science, understanding the lagged causal influence of greenhouse gas emissions on global temperature changes is critical for accurate modeling and prediction. ...
... The challenges become more pronounced when analyzing timeseries data, which introduce temporal complexities like lags, feedback loops, and seasonal or trending patterns [23]. Real-world timeseries rarely disclose the true underlying causal mechanisms, and their complexity can either obscure or mimic causality. ...
... Synthetic datasets with established causal relationships can fill this gap by providing controlled experimental environments [37,53]. However, many existing synthetic benchmarks are designed with assumptions that may not reflect real-world complexities [23]. In some cases, researchers tailor synthetic data to highlight the strengths of their proposed approaches, which can yield more favorable performance on these benchmarks [35]. ...
Robust causal discovery in time series datasets depends on reliable benchmark datasets with known ground-truth causal relationships. However, such datasets remain scarce, and existing synthetic alternatives often overlook critical temporal properties inherent in real-world data, including nonstationarity driven by trends and seasonality, irregular sampling intervals, and the presence of unobserved confounders. To address these challenges, we introduce TimeGraph, a comprehensive suite of synthetic time-series benchmark datasets that systematically incorporates both linear and nonlinear dependencies while modeling key temporal characteristics such as trends, seasonal effects, and heterogeneous noise patterns. Each dataset is accompanied by a fully specified causal graph featuring varying densities and diverse noise distributions and is provided in two versions: one including unobserved confounders and one without, thereby offering extensive coverage of real-world complexity while preserving methodological neutrality. We further demonstrate the utility of TimeGraph through systematic evaluations of state-of-the-art causal discovery algorithms including PCMCI+, LPCMCI, and FGES across a diverse array of configurations and metrics. Our experiments reveal significant variations in algorithmic performance under realistic temporal conditions, underscoring the need for robust synthetic benchmarks in the fair and transparent assessment of causal discovery methods. The complete TimeGraph suite, including dataset generation scripts, evaluation metrics, and recommended experimental protocols, is freely available to facilitate reproducible research and foster community-driven advancements in time-series causal discovery.
... Несмотря на широкую распространенность регрессионного анализа в современных исследованиях подобной тематики, основным методом нашей работы стал тест каузальности Гренджера. Выбор данного метода обоснован его способностью выявлять динамические зависимости во временны х рядах, а также широким применением для определения причинно-следственных связей между экономическими переменными (Granger, 1969). ...
... Что касается методов исследования, то в силу ограниченности данных применение регрессионного анализа было не обоснованно, и мы обратились к другим методам анализа, которые на сегодняшний день не столько распространены в литературе, но полезны для решения узких задач -тест причинности по Грэнджеру. Этот инструмент позволяет выявлять существование влияния между факторами и в отличие от регрессионного анализа, способен идентифицировать направленность такого влияния (Granger, 1969). ...
The study identifies external environmental factors that influence industrial enterprises’ innovation activity drawing on the example of China – the global leader in innovative development. The subject of this study is the measurement of innovation activity within the industrial sector, such as: research and development activity, the number of new product development projects and active patents for inventions, revenue from sales of new products and their exports. The selection of external success factors is based on literature review on relevant topics including: innovation infrastructure, institutions, cooperation and financing. Financing is employed as a control factor that has been repeatedly substantiated by other empirical studies. The analysis is based on official Chinese statistics (2011-2022), the Global Innovation Index sub-indices and data from the Statista platform. The time series sample is subjected to a detailed examination through descriptive statistics, correlation analysis and the calculation of an internal consistency coefficient. The Granger causality test is employed as a primary research method. The findings reveal that financing represents the most crucial element influencing the innovation activity of enterprises. The provision of startup financing and venture capital investment has been demonstrated to stimulate an increase in the number of innovative projects and new products. The study demonstrates that scientific collaboration, formation of clusters, and regulatory framework exert a considerable influence on the innovation processes of enterprises engaged in R&D. The empirical data prove that regulatory framework stimulates the number of active patents, and high patent activity in turn promotes cooperation and imports. The study thus asserts the importance of external factors, in particular financing and collaboration, which entail the success of innovation projects within the industrial sector. The findings are of significant importance for the development of new national development programmes and the adaptation of existing ones, as well as for the management of companies seeking to enhance the efficiency of their investment in innovation projects.
... 3. Temporal causal inference via Granger causality (Granger, 1969) applied to latent traits, emotional signals, and stance trajectories, allowing us to model how one agent's state causally influences another's uncertainty. ...
... Recent advances in causal inference have enabled deeper analysis of influence and information flow in social systems. Granger causality (Granger, 1969), in particular, has proven effective in time series settings for identifying whether one process provides statistically significant predictive information about another. Applications of Granger causality span neuroscience, econometrics, and social behaviour (Bach & Dolan, 2012), but have rarely been used in the context of deep learning for multi-agent interactions. ...
Human social behaviour is governed by complex interactions shaped by uncertainty, causality, and group dynamics. We propose Causal Spherical Hypergraph Networks (Causal-SphHN), a principled framework for socially grounded prediction that jointly models higher-order structure, directional influence, and epistemic uncertainty. Our method represents individuals as hyperspher-ical embeddings and group contexts as hyper-edges, capturing semantic and relational geometry. Uncertainty is quantified via Shannon en-tropy over von Mises-Fisher distributions, while temporal causal dependencies are identified using Granger-informed subgraphs. Information is propagated through an angular message-passing mechanism that respects belief dispersion and directional semantics. Experiments on SNARE (of-fline networks), PHEME (online discourse), and AMIGOS (multimodal affect) show that Causal-SphHN improves predictive accuracy, robustness, and calibration over strong baselines. Moreover, it enables interpretable analysis of influence patterns and social ambiguity. This work contributes a unified causal-geometric approach for learning under uncertainty in dynamic social environments .
... Here, we present two families of causal inference methods: those based on Granger's concept of causality [150], and those based on information theory and the concept of Figure 6: (a) Temporal evolutions of two signals ξ 1 (t) and ξ 2 (t), and the influence of previous information about ξ 2 on estimation of the future state of ξ 1 (assuming that there is in fact causation from ξ 2 to ξ 1 ); (b) primary causal interactions between the large-scale coherent structures observed in flow past a square cylinder [152]; (c) diagram of causal inference model based on a decomposition of the flow field on its informative and residual components (top), and implementation of this decomposition for the prediction of wall shear stresses and the development of an opposition control for drag reduction in a turbulent channel (bottom) [153]. Reprint permissions granted by Cambridge University Press. ...
... Causal inference methods based on Granger's concept of causality build upon the statistical approach [150]. In this framework, causality is measured as the improvement in the prediction of a temporal variable ξ 1 (t) ∈ R n , before and after incorporating past values of another variable ξ 2 (t) ∈ R n . ...
A wide range of techniques exist for extracting the dominant flow dynamics and features about steady, or periodic base flows. However, there have been limited efforts in extracting the dominant dynamics about unsteady, aperiodic base flow. These flows appear in many applications such as when there is a sudden change in the flow rate through a pipe, when an airfoil experiences stall, or when a vortex forms. For these unsteady flows, it is valuable to know not only the dynamics of the base flow but also the features that form around this base flow. Here, we discuss the current state of research on extracting important flow structures and their dynamics in such cases with time-varying base flows. In particular, we consider data-driven decompositions, operator-based methods, causality analysis, and some other approaches. We also offer an outlook and call attention to key areas that require future efforts.
... Granger causality analysis is used to test whether the information in one variable series can be estimated using the information in another variable series. Granger (1969) explains as an exogenous variable and as an endogenous variable if is the cause of . The F-statistic is used to test whether the hypothesis 0 , which states that there is no causality between the variables, is accepted (Zelka & Yıldırım, 2022). ...
... In this study, the Granger (1969) causality test, the Fourier Toda-Yamamoto causality test (Enders & Jones, 2015), the Fourier Standard Granger causality test (Enders & Jones, 2015), the Fourier Toda & Yamamoto test -cumulative frequency (Nazlioglu et al., 2016) and the Hatemi-J (2012) causality analyses were employed to investigate the causal relationship between the Airline Price Index (AIR) and Brent Crude Oil (OIL) and the Dollar Index (DXY). In the context of causality analysis, it is essential to ascertain whether the series in question are integrated at the same level. ...
The Airline Price Index is a stock market index that monitors the stock prices of companies within the airline industry, reflecting the overall value of these airline companies. This index is an important tool for assessing the financial health, economic status, and market performance of the airline industry. This study aims to explore the causal relationships between the Airline Price Index (AIR), the Dollar Index (DXY), and Oil Prices (OIL), thereby contributing to a deeper understanding of the dynamics influencing the airline industry. In this study, a number of different Granger causality tests are employed, including the Granger causality test, the Fourier Toda-Yamamoto causality test, the Fourier Standard Granger causality test, and the Fourier Toda & Yamamoto test, as well as the cumulative frequency test and the Hatemi-J (2012) causality analysis. These are used to examine the causality relationship between AIR, OIL and DXY. The results of the analysis indicate that the expected causality relationship from OIL to AIR is not supported by the entire analysis method. In contrast, the Granger causality test results indicate that there is a unidirectional causal relationship from AIR to OIL and DXY. Furthermore, the results of the Fourier-Toda-Yamamoto and Fourier standard Granger analyses are consistent with one another. Considering these findings, it can be concluded that there is bidirectional causality between DXY and AIR. The findings of the Fourier-Toda & Yamamoto (cumulative frequency) analysis indicate the presence of a causality relationship from AIR to OIL and DXY. These findings are consistent with the results of Granger causality tests.
... One of the first and most robust empirical analyzes on the ABCT is produced by Wainhouse (1984) showing that monetary policy affects interest rates and consequently impacts the output, through the causality test Granger (1969) and using monthly US data for the period from 1959 to 1981, empirically find strong evidences of the austrian business cycle theory. ...
O objetivo deste artigo é avaliar os efeitos da política monetária sobre o ciclo econômico e a taxa de inflação por meio de variações nos preços relativos da economia no Brasil, com base em séries temporais mensais para o período de março de 2007 a setembro de 2017. A econométrica os resultados obtidos pelo GMM corroboram com a teoria austríaca dos ciclos econômicos e mostram que a oferta de moeda não afeta apenas os preços relativos da economia, mas também indiretamente o hiato do produto (ciclos econômicos) e a inflação no curto prazo. Considerando também que uma expansão monetária contribui para a expansão do crédito e que são fortes as evidências de implicações microeconômicas decorrentes de uma variação na oferta de moeda, esses resultados tornam-se relevantes para avaliar os efeitos das políticas monetária e de crédito sobre o nível de atividade econômica e a taxa da inflação, por meio de variações nos preços relativos da economia.
... The direction of the causation for the relationship of variables will be tested using a causality test (Sadorsky, 2009;Apergis and Payne 2010a;Al-Mulali, 2014;Pao et al., 2014;Jebli and Youssef, 2015;Kahia et al., 2017). The Granger Causality Wald test (Granger, 1969) will be adopted to determine if there is a unidirectional, bidirectional causality or no causality between two variables. The short run dynamic between variables is being identified within the cointegrated model (Engle and Granger, 1987). ...
Despite Malaysia's long-standing efforts to diversify its energy sources and increase renewable energy consumption since 1980, the progress has been hindered by inefficiencies and technological limitations, while the economic impact of these changes remains unmeasured. This study, using annual World Bank data from 1990 to 2022, examines the relationship and impact of renewable energy consumption on Malaysia’s economic and industry, service, and agriculture sectors’ growth. The findings suggest that the renewable energy sector is cointegrated and has a significant positive impact on the overall economy, and the industry and service sectors in the long run. Furthermore, the Granger causality Wald test results support the growth, neutrality, and conservation hypotheses for the overall economy, industry, agriculture, and service sectors, respectively, in the short run. Overall, the results suggest that renewable energy could benefit Malaysia's economy in varying degrees across the three important sectors. Therefore, prioritizing investments in renewable energy consumption within the industry and service sectors is crucial, as they offer the largest positive impact on economic growth. Additionally, the adoption of technologies in agricultural sector is low, thus efforts to enhance access to modern farming techniques using renewable energy should be promoted to improve the performance of the agricultural sector.
... Bu test, bir değişkenin geçmiş değerlerinin, diğer değişkenin gelecek değerlerinin tahmin edilmesinde istatistiksel olarak anlamlı bir etkiye sahip olup olmadığını sınamaktadır. Bunun için, bu analizde Vektör Otoregresyon (VAR) modeli kullanılır (Granger, 1969). Düzeyde durağan olmayan (I(1)) seriler için ise kısa ve uzun dönem ilişkiler Gecikmesi Dağıtılmış Otoregresif Model (ARDL) ile analiz edilecektir (Pesaran, vd, 2001). ...
Bu çalışma, Türkiye'de kredi hacmi ve ekonomik büyüme arasındaki ilişkiyi incelemektedir. Araştırma, 2007:4-2023:4 dönemini kapsayan üçer aylık verileri kullanarak gerçekleştirilmiştir. Analizde, finansal gelişme göstergesi olarak banka kredileri, ekonomik büyüme göstergesi olarak ise GSYİH kullanılmıştır. Çalışmada hem kısa hem de uzun dönemli ilişkiler incelenmiştir. Kısa dönemli ilişki VAR modeli tabanlı Granger nedensellik testi ile analiz edilirken, uzun dönemli ilişki ARDL sınır testi yaklaşımıyla değerlendirilmiştir. Bu metodolojik çeşitlilik, analizin derinliğini artırmaktadır. Araştırma bulguları, kredi büyümesi ve ekonomik büyüme arasında anlamlı bir ilişkinin varlığını ortaya koymaktadır. Granger nedensellik testi sonuçları, değişkenler arasında kısa dönemli bir ilişkinin mevcut olduğunu gösterirken, ARDL sınır testi sonuçları uzun dönemli bir ilişkinin varlığına işaret etmektedir. Çalışma, 2008-2009 küresel finansal krizi ve 2020 COVID-19 pandemisi gibi önemli ekonomik olayların hem kredi büyümesi hem de ekonomik büyüme üzerindeki etkilerini de incelemiştir. Bu dönemlerde her iki değişkende de keskin düşüşler gözlemlenmiştir. Sonuç olarak, bu araştırma Türkiye'de finansal gelişme ve ekonomik büyüme arasındaki karmaşık ilişkiyi anlamak için önemli öngörüler sunmaktadır. Elde edilen bulgular, politika yapıcılar ve araştırmacılar için değerli bilgiler sağlamakta ve gelecekteki ekonomik politikaların şekillendirilmesinde yol gösterici olabilecek niteliktedir.
... The study employs the linear Granger (1969) causality test in the VECM theme, to examine the short-run and long-run linearity relationship among the variables in bivariate and multivariate mode. To provide accuracy in the estimate of the relationship, it is thus necessary to prior determine the presence of unit root and cointegration between the time series. ...
This study examines the impact of the monetary policy rate (MPR) on the interbank market rate (INTB) in Ghana using monthly data from January 2000 to December 2024. Employing the Johansen cointegration test and Vector Error Correction Model (VECM), the study finds a stable long-run relationship between MPR and INTB, with the interbank rate adjusting to policy changes over time. However, short-run adjustments are sluggish, indicating inefficiencies in the transmission process. Market liquidity constraints, asymmetric information, and structural rigidities contribute to these delays. The findings suggest that while monetary policy remains an effective tool for influencing short-term market rates, enhancing interbank market efficiency is crucial for improving policy transmission. Policy recommendations include strengthening liquidity management, reducing market segmentation, improving policy signaling, and leveraging financial technology to enhance interbank rate responsiveness. These measures can improve the effectiveness of monetary policy in Ghana, ensuring better financial stability and economic growth.
... Following the coefficient estimates, the causal relationship between the variables should be investigated. At this stage, Granger (1969) test was used. Granger's basic proposition is as follows; Equation (4) has two variables, r and l, that do not change over time. ...
We use the PVAR econometric method to analyze the main dynamics of sustainable growth, namely environmental pollution, growth and energy consumption, for the ASEAN panel from 1990 to 2020. Energy consumption and technology are important factors for economic growth, but they also have environmental implications. Therefore, we consider how environmental factors and information and communication technology (ICT) affect gross domestic product per capita (GDPPC), renewable energy consumption (REN) and carbon dioxide emissions (CO2). Our results show that GDPPC has a negative impact on both Ecological Footprint (EF) and ICT, but EF and ICT do not significantly influence GDPPC. Moreover, EF has a positive and significant effect on ICT, but not vice versa. ICT also positively affects REN and negatively affects CO2, while EF does not have a significant impact on these variables. Thus, we conclude that for ASEAN countries, enhancing EF can improve technological development and environmental quality.
... In order to apply Granger Causality Test, the time series to be used must be stationary. For this reason, it should first be checked whether the series are stationary (Granger, 1069). The commonly used stationarity tests in the literature are the Enhanced Dickey-Fuller (ADF) unit root test developed by Dickey and Fuller (1981) and the Philips Perron (PP) unit root test. ...
The main aim of this study is to determine the causal relationship between Bitcoin, the cryptocurrency with the highest market volume, and the BIST100, which represents the Turkish stock market. The study period is divided into two temporal periods considering the COVID-19 pandemic, which affected all markets and caused significant changes in their performance. The time intervals are defined as pre-pandemic (18/01/2017 - 9/03/2020) and pandemic period (10/03/2020 -05/05/2023). Daily data was used for analysis, and weekend prices were excluded from the analysis as the stock market operates only on weekdays. The Vector Autoregressive (VAR) model-based Granger Causality Test was used to examine the data. The results of the analysis indicated that no causality could be detected between Bitcoin and BIST100, and vice versa, during both the pre-pandemic and pandemic periods.
... Também foi estimado o teste de Causalidade de Granger, que mensura a causalidade do passado para o presente, entre os pares de variáveis, conforme Granger (1969). E o teste de cointegração de Johansen, que de acordo com Johansen (1988), busca identificar se existe uma relação de longo prazo entre as variáveis do modelo. ...
Resumo A discussão sobre a inflação dos alimentos é de grande relevância no debate econômico, pois para além da questão relacionada a oscilação de preços, este problema afeta diretamente a capacidade da população em acessar uma alimentação básica e, consequentemente, garantir a segurança alimentar. Assim, o presente trabalho teve como objetivo quantificar e discutir os possíveis efeitos que o controle direto da oferta do arroz, do feijão e do trigo via estoques reguladores, pode gerar sobre o nível de preço destes alimentos. Para tanto, foi criado um índice de preços para os alimentos citados e o método principal de estimação utilizado foi o de Vetores Autorregressivos com Correção de Erros. Os resultados obtidos mostram que os efeitos da operacionalização da oferta do arroz têm influência sobre o índice de preços criado, em períodos subsequentes ao uso da medida. Já os preços do feijão e do trigo demonstraram ser significativamente menos sensíveis a operacionalização da sua oferta, via estoques, por exemplo. Assim, os resultados sugerem que não se pode descartar a hipótese de que a capacidade de operacionalizar a oferta de produtos básicos, via estoques reguladores, pode proporcionar um bom auxílio à política monetária, no objetivo de estabilizar os preços destes itens.
... The Quantile Granger Causality Test, developed from Granger's theory [42], was extended by Koenker and Xiao [43] to analyze causal relationships at the mean level and across different distribution quantiles. This approach allows for a more comprehensive understanding of how causality varies at various points in the distribution, providing insights into the dynamics that may not be visible through average effects alone. ...
... The conditioning event is treated as a natural experiment, and the evolution of the actual system is substituted by a data-driven model. These methods were pioneered by Granger [116] in economics, and have become very popular because of their savings with respect to strict causality, but an analysis of the possible pitfalls is given in [107], and they share with all data-driven methods the caveats raised in our previous section on the relation between a system and its attractor. An application to fluid mechanics that addresses some of these drawbacks is [117], and it is interesting that their results regarding inner-outer interactions in wall-bounded turbulence are compatible with those of the interventional experiments in [103], in spite of the very different methodologies. ...
This paper is a personal overview of the efforts over the last half century to understand fluid turbulence in terms of simpler coherent units. The consequences of chaos and the concept of coherence are first reviewed, using examples from free-shear and wall-bounded shear flows, and including how the simplifications due to coherent structures have been useful in the conceptualization and control of turbulence. It is remarked that, even if this approach has revolutionized our understanding of the flow, most of turbulence cannot yet be described by structures. This includes cascades, both direct and inverse, and possibly junk turbulence, whose role, if any, is currently unknown. This part of the paper is mostly a catalog of questions, some of them answered and others still open. A second part of the paper examines which new techniques can be expected to help in attacking the open questions, and which, in the opinion of the author, are the strengths and limitations of current approaches, such as data-driven science and causal inference.
... While (5) allows Y s to influence X t for s < t, (6) rules out such feedback (we allow dependence on the initial condition throughout). Restriction (6) is a counterpart of Granger's (1969) definition of "Y does not cause X" in the panel data setting with latent heterogeneity. If X t is a choice variable, and Y t−1 is a component of the agent's begining-of-period t information set, then (6) typically implies strong restrictions on economic behavior (e.g., Ashenfelter and Card, 1985) and/or the structure of agent's information sets (e.g., Chamberlain, 1984Chamberlain, , 1985. 4 Our approach, by accommodating unrestricted feedback and heterogeneity, allows the researcher to proceed without maintaining such assumptions. ...
Many panel data methods, while allowing for general dependence between covariates and time-invariant agent-specific heterogeneity, place strong a priori restrictions on feedback: how past outcomes, covariates, and heterogeneity map into future covariate levels. Ruling out feedback entirely, as often occurs in practice, is unattractive in many dynamic economic settings. We provide a general characterization of all feedback and heterogeneity robust (FHR) moment conditions for nonlinear panel data models and present constructive methods to derive feasible moment-based estimators for specific models. We also use our moment characterization to compute semiparametric efficiency bounds, allowing for a quantification of the information loss associated with accommodating feedback, as well as providing insight into how to construct estimators with good efficiency properties in practice. Our results apply both to the finite dimensional parameter indexing the parametric part of the model as well as to estimands that involve averages over the distribution of unobserved heterogeneity. We illustrate our methods by providing a complete characterization of all FHR moment functions in the multi-spell mixed proportional hazards model. We compute efficient moment functions for both model parameters and average effects in this setting.
... Building on the knowledge of multi-frequency quantile regression (MFQR) above, this study employs the multi-frequency quantile Granger causality (MFQGC) approach suggested by Adebayo [13]. While Özkan et al. 's [31] wavelet quantile Granger causality (WGQC) combined Granger causality [32] with the maximal overlapping discrete wavelet transform (MODWT) by Percival and Walden (2000) and the quantile method of Li et al., [33], the MFQGC integrates Granger causality with EEMD Wu & Huang [34] and the quantile method of Li et al., [33]. The MFGQC is defined as follows: ...
This study pioneers an investigation into the impact of policy uncertainty (climate and economic) and industrial production on renewable energy demand, focusing on the United States. Utilizing monthly data spanning from April 1987 to September 2024, the research employs advanced multi-frequency quantile regression and multi-frequency quantile Granger causality to analyze these dynamics across various frequencies and quantiles. These advanced methodologies can capture associations between variables across multiple frequencies and quantiles. The findings reveal that industrial production and economic growth consistently exert a positive influence on renewable energy demand across all quantiles and frequencies. Conversely, policy uncertainty (both climate and economic) demonstrates mixed effects on renewable energy demand. Aligned with Sustainable Development Goals (SDGs), particularly SDG 7, 8, and 13, this study highlights the critical need for balanced strategies.
... Section "Conclusions" concludes. Granger's (1969) non-causality test is a statistical hypothesis test whether lagged values of a variable or variables contain additional information to better predict an other variable. Technically speaking, X Granger-causes Y if the lagged values of X are statistically significant predictors of Y , while controlling for the lagged values of Y. ...
I test for Granger non-causality between green and conventional (plain vanilla) bond yields/prices in the secondary market for each domestic sovereign green bond outstanding in September 2023. Results dominantly show no causality, which–in light of the co-movement of the related time series–suggests the non-stable motives of the green investors. To complement the causality analysis, I also examine possible cointegrating relationships. These results show that long-term links are mostly highly unstable, which also indicates investor ‘flexibility’ rather than a stable green commitment. I argue that the presented non-structural approach offers a practical screening tool to support further case-by-case structural analyses or financial engineering per se.
... Toutefois, l'analyse de la causalité de Granger (1969) donne des résultats intéressants qui peuvent aider dans la conception des stratégies de réalisation des villes intelligentes en Afrique. Le constat est que le niveau de développement économique détermine la qualité de l'air. ...
Ce papier examine les stratégies de développement des villes intelligentes pour ramener la croissance en Afrique sur la période de 2010-2020. Les résultats de la recherche fréquentiste montrent que les pertes économiques exercent une pression négative. Par ailleurs, la pression négative qu'exerce l'envol inflationniste participe à la vulnérabilité sociale. Pendant que le recul des dépenses publiques freine la relance économique. Contrairement, la diminution des décès pendant les catastrophes et la diminution du retard économique exercent une pression positive. On ne peut pas ignorer l'existence d'une relation unidirectionnelle de la croissance à la qualité de l'air. Du fait léthargique de certains pays africains à la réalisation des stratégies de développement des villes, il est difficile dans ce contexte de ramener pleinement la croissance en Afrique. Si l'élaboration de plans novateurs semble intéressante pour parvenir à un meilleur jour, elle interpelle également l'Union Africaine à intégrer la stratégie de villes intelligentes dans la réalisation de la vision 2063 pour un développement accéléré des pôles urbains de croissance en Afrique. Mots-Clés : Ville intelligente, développement durable, attractivité, croissance, Afrique.
Abstract This paper examines smart city development strategies to bring growth back to Africa over the period 2010-2020. The results of frequentist research show that economic losses exert a negative pressure. In addition, the negative pressure exerted by the inflationary boom contributes to social vulnerability. Meanwhile, the decline in public spending is slowing down the economic recovery. On the other hand, the decrease in deaths during disasters and the decrease in economic backwardness are putting positive pressure on. We cannot ignore the existence of a unidirectional relationship between growth and air quality. The lethargic nature of some African countries in implementing urban development strategies makes it difficult in this context to bring growth fully back to Africa. While the development of innovative plans seems interesting to achieve a better day, it also calls on the African Union to integrate the smart cities strategy into the realization of the 2063 vision for an accelerated development of urban growth poles in Africa.
... The key point is that deposit taking and on lending by banks to economic sectors (as part of their intermediation role) mirror a dynamic banking business. In that regard, the adequacy of deposit insurance fund in Sierra Leone should be some percentage of the country's national cake (GDP) consistent with Sierra Leone broad macro financial and economic framework [36][37][38][39][40]. ...
Deposit protection funds are now conventional across sovereigns as integral to financial system safety net and supportive to financial system stasis. This paper examines the potential adequacy of deposit protection fund in Sierra Leone by employing scenario analysis with data vintage 2021Q1-2023Q4 regarding the banking ecosystem health check. The paper unravels that adequacy of deposit protection fund in Sierra Leone should mimic some proportion of the gross domestic products (GDP) entrenched on Sierra Leone macroeconomic and financial target framework that is dynamic. Thus, the paper recommends the need for robust banking supervision and macroeconomic and financial stability assessments can assuage unintended ramifications of deposit protection fund on banking ecosystem risk in buoyant episodes, thereby enthroning suitable incentive framework is crucial for financial system stasis in Sierra Leone.
... A common challenge in causal learning is unobserved common causes. In the present context, if the common cause of two nodes is removed due to lower temporal resolution, then there will be bidirectional correlations between nodes, as described in Section 2. Many causal structure learning methods produce bidirected edges [Runge et al., 2019;Granger, 1969;Lütkepohl, 2005], but methods that do not account for undersampling will not produce bidirected edges [Gates and Molenaar, 2010;Barnett and Seth, 2009;Bressler and Seth, 2003;Sanchez-Romero et al., 2019a]. We present an approach where our method can be used as a meta-solver, thereby gaining the benefits of the chosen first-order method, while accommodating undersampling due to our method. ...
Learning graphical causal structures from time series data presents significant challenges, especially when the measurement frequency does not match the causal timescale of the system. This often leads to a set of equally possible underlying causal graphs due to information loss from sub-sampling (i.e., not observing all possible states of the system throughout time). Our research addresses this challenge by incorporating the effects of sub-sampling in the derivation of causal graphs, resulting in more accurate and intuitive outcomes. We use a constraint optimization approach, specifically answer set programming (ASP), to find the optimal set of answers. ASP not only identifies the most probable underlying graph, but also provides an equivalence class of possible graphs for expert selection. In addition, using ASP allows us to leverage graph theory to further prune the set of possible solutions, yielding a smaller, more accurate answer set significantly faster than traditional approaches. We validate our approach on both simulated data and empirical structural brain connectivity, and demonstrate its superiority over established methods in these experiments. We further show how our method can be used as a meta-approach on top of established methods to obtain, on average, 12% improvement in F1 score. In addition, we achieved state of the art results in terms of precision and recall of reconstructing causal graph from sub-sampled time series data. Finally, our method shows robustness to varying degrees of sub-sampling on realistic simulations, whereas other methods perform worse for higher rates of sub-sampling.
... The formalization of feedback is typically ascribed toGranger (1969), whileEngle, Hendry, and Richard (1983) provide an early rigorous distinction between weak and strict exogeneity. See alsoSims (1972),Chamberlain (1982) for further discussions and an empirical example. ...
This paper studies linear time‐series regressions with many regressors. Weak exogeneity is the most used identifying assumption in time series. Weak exogeneity requires the structural error to have zero conditional expectation given present and past regressor values, allowing errors to correlate with future regressor realizations. We show that weak exogeneity in time‐series regressions with many controls may produce substantial biases and render the least squares (OLS) estimator inconsistent. The bias arises in settings with many regressors because the normalized OLS design matrix remains asymptotically random and correlates with the regression error when only weak (but not strict) exogeneity holds. This bias' magnitude increases with the number of regressors and their average autocorrelation. We propose an innovative approach to bias correction that yields a new estimator with improved properties relative to OLS. We establish consistency and conditional asymptotic Gaussianity of this new estimator and provide a method for inference.
... Zaman serilerinin eğilimlerinin değerlendirilmesinin yanı sıra, hem tüm uçak trafiğinin parametreler üzerindeki etkisi hem de parametrelerin birbiri üzerindeki etkilerinin belirlenmesi için Granger Nedensellik testi zaman serilerine uygulanmıştır. Daha çok ekonomi ve sosyal bilimler (Takım 2010, Shojaie ve Fox 2022) alanlarında kullanılan test,Granger (1969) tarafından önerilen istatistiksel bir hipotez testidir. Bir değişkenin diğer değişkenin tahmininde etkili olup olmadığını anlamaya yardımcı olan güçlü bir istatistiksel araç olarak anılmaktadır. ...
Havalimanları toplumun ulaşım ihtiyacının önemli bir kısmını karşılamaktadır. Ancak, kentleşmenin artışı ile yeni havalimanlarının inşaatı için elverişli alanlar azalmış, şehre uzak noktalarda rehabilite edilmiş alanlara veya denize dolgu alanlara inşa edilmeye başlanmıştır. Ancak doğal ekosistemin tahrip edilmesiyle inşa edilen bu yapıların, çevreye verebileceği olası zararların izlenmesi gerekmektedir. Bu bağlamda izlenmesi gereken çevresel etkilerden biri, yeni inşa edilen bu havalimanlarının etkilediği hava kalitesidir. Uzaktan algılama kaynakları, yer istasyonları gibi nokta bazlı bir veri kaynağı olmamakla birlikte belirli periyotlarda veri sağlayarak mekânsal ve zamansal analizlere olanak sağlamaktadır. Çalışma kapsamında, ülkemizde en son faaliyete açılan İstanbul Havalimanı ve Rize-Artvin Havalimanı üzerinde hava kalite parametrelerinde meydana gelen değişimler uydu kaynaklarından elde edilen aylık zaman serileri ile incelenmiştir. Sentinel-5P uydusu ile ultraviyole aerosol indeks (UV), karbon monoksit (CO), azot dioksit (NO2), ozon (O3) ve kükürt dioksit (SO2) hava kalite parametrelerinin 2018-Mayıs 2024 periyodu için zaman serileri Google Earth Engine bulut platformu kullanılarak elde edilmiştir. Parametre bazında incelenen zaman serisi analizlerinde, UV, SO2, NO2 ve O3 parametrelerinde artan bir eğilim tespit edilirken, CO parametresinde stabil bir eğilim görülmüştür. Gerçekleştirilen istatistiksel analizler doğrultusunda uçak trafiği verileri ile CO ve NO2 parametreleri arasında pozitif korelasyonların (havalimanları özelinde sırasıyla CO parametresi ile 0.48 ve 0.41, NO2 parametresi ile 0.66 ve 0.73) yanı sıra istatistiksel olarak anlamlı sonuçlar elde edilmiştir.
... This study examines the relationship between globalisation, its subdimensions and economic growth. Panel Granger approach [70] is utilised to evaluate the relationships between the variables which is highly preferred by many under similar discussions [63,71,72]. Prior to conducting the causality test, Cross-sectional dependency, stationarity, homogeneity was tested. ...
Globalisation is recognised as a prospective dynamic that facilitates the performance and expansion of economies. This study analyses the causal progression between globalisation, its sub dimensions (economic, social and political) and economic growth spanning 97 countries and six regions (Africa, Asia, Europe, North America, Oceania, and South America) covering the period from 1971 to 2021. The Panel Granger causality test is employed as the statistical methodology to comprehend the nexus between globalisation and economic growth. The Granger results reveal bi-directional causal flows between economic growth and globalisation in Asia, North America, and Oceania, along with one-way causal flows in Africa, South America, and Europe. Bidirectional dynamics pertaining to economic globalisation were also revealed in Asia, Africa, Oceania, and Europe. This study recommends the enhancement of regional integration, addressing of structural changes, leveraging the use of technology, and the development of comprehensive globalisation strategies with respect to regions with the intention of reinforcing their globalisation-growth stance, while complementing the Sustainable Development Goals of the United Nations.
... Como resultado, uma vez que as 6 variáveis cointegram, pode-se realizar testes de causalidade, seguidos de modelos empíricos. Com base nos testes de Granger (1969), pode-se testar a relação de causalidade entre a variável dependente e as demais variáveis explanatórias. Inicialmente deve-se verificar o número ótimo de lags, conforme tabela 8 apresentada a seguir. ...
Este estudo investiga o impacto da Grande Depressão sobre o consumo na economia dos Estados Unidos, no período de 1929:04 a 1936:04. Os resultados empíricos mostram que as variáveis exploratórias — renda pessoal, índice Dow Jones, moeda em circulação, número de falências nos negócios e total de exportações — podem explicar o comportamento do consumo. Destacam-se contribuições como: o uso do consumo, em vez da produção, como variável dependente; o índice Dow Jones como causa de Granger do consumo; as variáveis exploratórias a preços correntes afetando a variável dependente a preços constantes; e, por fim, a complementaridade entre as explicações de Friedman e Keynes sobre a Grande Depressão.
... Causal inference has been studied by many scientists over the years [32][33][34]40]. It has been observed through a number of experiments that causal relationships can be discovered from observational data [15,34,39]. A possible approach is through a randomized controlled trial. ...
Discovering causal relationships from observed data has applications in many fields including Bioinformatics. The PC (Peter and Clarke) algorithm and its parallel version are the state-of-the-art methods for causal discovery. However, the high runtime of these algorithms makes them inefficient for high-dimensional data. For example, to discover causal regulatory expression in gene, the number of nodes (factors) can easily be more than 2000. For such an instance, even the fastest variant of PC today would take days to complete. To solve this, we propose a recursive parallel causal discovery algorithm (RPCD). RPCD constructs the graph recursively, which leads to significant savings in the number of conditional tests. We evaluated RPCD on a number of causal network datasets as well as on a real-life incident dataset. The experimental results show that it improves both the running time and accuracy metrics significantly.
Graphical Abstract
... 1. GC (Granger 1969). Granger Causality (GC) is a method that incorporates statistical testing to determine the causality between two time series. ...
The rise of social media has been accompanied by a dark side with the ease of creating fake accounts and disseminating misinformation through coordinated attacks. Existing methods to identify such attacks often rely on thematic similarities or network-based approaches, overlooking the intricate causal relationships that underlie coordinated actions. This work introduces a novel approach for detecting coordinated attacks using Convergent Cross Mapping (CCM), a technique that infers causality from temporal relationships between user activity. We build on the theoretical framework of CCM by incorporating topic modelling as a basis for further optimizing its performance. We apply CCM to real-world data from the infamous IRA attack on US elections, achieving F1 scores up to 75.3% in identifying coordinated accounts. Furthermore, we analyse the output of our model to identify the most influential users in a community and uncover leader-follower dynamics based on inferred causal relationships. We also demonstrate how our method reveals coordinated behaviour across different time periods, including campaigns predating the 2016 elections. We apply our model to a case study involving COVID-19 anti-vax related discussions on Twitter. Our results demonstrate the effectiveness of our model in uncovering causal structures of coordinated behaviour, offering a promising avenue for mitigating the threat of malicious campaigns on social media platforms.
... The Causal Markov and Causal Faithfulness assumptions have formal definitions requiring technical notation that are beyond the scope of this article. For a full discussion of these assumptions, we refer the reader to Pearl (2000) 23 and Spirtes and Zhang (2016) 60 . ...
In ecology, causal questions are ubiquitous, yet the literature describing systematic approaches to answering these questions is vast and fragmented across different traditions (e.g., randomization, structural equation modeling, convergent cross mapping). In our Perspective, we connect the causal assumptions, tasks, frameworks, and methods across these traditions, thereby providing a synthesis of the concepts and methodological advances for detecting and quantifying causal relationships in ecological systems. Through a newly developed workflow, we emphasize how ecologists’ choices among empirical approaches are guided by the pre-existing knowledge that ecologists have and the causal assumptions that ecologists are willing to make.
... Financial variables often have their predicted correlations examined using Granger causality tests. Time series analysis, which Granger (1969) popularised, has found extensive use in economics and finance ever since. For example, Ewing and Malik (2013) used Granger causality tests to analyze the relationship between oil futures prices and stock market volatility, finding evidence of bidirectional causality. ...
This study examines the inter-relation between WTI crude oil futures prices and two energy-related exchange-traded funds (ETFs): the ETF of iShares Global Clean Energy (Clean Energy) and the ETF of Energy Sector SPDR Fund (Traditional Energy). Using Granger causality tests and the Dynamic Conditional Correlation Generalized Autoregressive Conditional Heteroskedasticity (DCC-GARCH) model, we analyze causal relationships and volatility transmission between these assets. The Granger causality results show that traditional energy markets dominate, while clean energy markets are becoming more influential on crude oil futures prices. Clean energy markets Granger-causes crude oil futures prices while traditional energy markets strongly Granger-causes crude oil futures. Clean energy ETFs have lower volatility persistence than conventional energy ETFs, according to the DCC-GARCH data, which also demonstrate time-varying correlations. Important insights for sustainable investment, energy policy, and risk management in the context of the world's energy transition are provided by these results, which emphasize the monetary interdependencies between energy ETFs and crude oil futures prices. Keywords: Crude oil futures prices, Clean energy ETFs, Traditional energy ETFs, Granger causality, DCC-GARCH, Sustainable Investing
... We made use of Dumitrescu and Hurlin (2012) causality test, which examines two primary facets of heterogeneity: the causal relationship's heterogeneity and the regression model's heterogeneity. This test is considered superior to the traditional Granger (Granger 1969) factor test for identifying directional effects. ...
Carbon emissions are a primary driver of environmental degradation and pose noteworthy risks to global efforts in combating climate change, particularly in ASEAN economies. While CO2 emissions are central to understanding environmental impacts, they represent just one aspect of the broader sustainability framework. This study examines the roles of eco‐innovation, green energy, globalization, economic growth, industrialization, and environmental taxes in reducing CO2 emissions within ASEAN economies from 1990 to 2022. Utilizing a cross‐sectional ARDL model, we explore the correlations among these factors, with special attention to the environmental dimension of sustainable development. Our findings highlight the negative relationships between eco‐innovation, green energy, industrialization, environmental taxation, and CO2 emissions, which contribute significantly to environmental sustainability both immediately and later on. Additionally, the study reveals reciprocal causal links between these elements and the emissions of CO2, offering critical policy insights for advancing environmental goals within the structure of the sustainable development goals (SDGs), especially those about climate action (SDG 13), affordable and clean energy (SDG 7), and sustainable industrialization (SDG 9). These insights provide a pathway for policy interventions that target CO2 mitigation while recognizing the diverse aspects of sustainable development.
... Based on above principles, testing the causal relationship between the five variables can assist in formulating more targeted measures to curb carbon emissions associated with farming activities. Thus, the VECM-based Granger causality method introduced by Engle and Granger is applied to inspect the interaction between the above parameters in this paper (Granger, 1969;Engle and Granger, 1987). The VECM can be written as following equation ( where ∆ represents the first-difference operator; −1 denotes the lagged error correction term; and represents the random error term. ...
... Granger causality was first employed by Granger (1969), and since then, this seminal approach has been cited broadly. Moreover, Dumitrescu and Hurlin (2012) have introduced renowned estimation causality appropriate to estimating a model with cross-section dependence (CSD) and heterogeneity. ...
Addressing the urgent global challenge of climate change and the pressing need for carbon neutrality, this research investigates how these mechanisms can drive sustainable development while balancing economic and environmental priorities. The current study assesses the role of green trading, environmental taxes, and eco-friendly technologies in promoting sustainable growth across 35 Organization for Economic Corporation and Development (OECD) countries from 2000 and 2021. Employing the Granger non-causality approach by Juodis et al. (Empir Econ 60:93–112, 2021), our analysis finds that these measures have significantly curbed pollution, though with varied impacts across income levels. Both high- and low-income economies show causal links between these policies and pollution reduction, emphasizing that thoughtful design, effective implementation, and continuous evaluation are essential for achieving long-term sustainability goals.
Graphical Abstract
... The Granger causality test (Granger, 1969) is used to determine whether one time series can forecast or predict another. The idea of forecasting or prediction is more relevant than testing whether Y causes or Granger-causes X, as is commonly asserted in the literature. ...
The integration of artificial intelligence (AI) in higher education is transforming students’ learning experiences, decision-making, and academic efficiency. This study explores student perceptions of AI’s benefits, challenges, and its role across academic disciplines. Findings indicate that while students recognize AI as a valuable educational tool, they also express concerns regarding privacy, reliance, and the need for improved AI training. Statistically significant differences were observed in AI’s impact on learning efficiency, decision-making, and academic engagement, supporting the hypothesis that AI enhances education but presents notable challenges. The study also highlights the necessity for balanced AI integration, ensuring that AI complements rather than replaces critical thinking and independent learning. These insights provide valuable implications for educators and institutions in developing AI policies that optimize learning outcomes while addressing numerous concerns.
Accurately predicting 5G bandwidth in vehicular mobility scenarios is essential for optimizing real-time communication (RTC) systems in a range of emerging vehicular applications, such as autonomous driving, in-car video conferencing, vehicular augmented reality (AR), and remote driving in dynamic urban environments. In this paper, we propose Prophet, a causality-aware Transformer model designed to forecast 5G throughput by capturing the complex causal relationships between control-plane 5G network events (e.g., handovers) and environmental factors (e.g., signal strength). Our key contributions include: (1) a wavelet-based encoder that efficiently captures long-term trends through multi-scale signal decomposition; (2) a control-plane causality attention mechanism in the decoder that focuses on localized, short-term throughput variations triggered by network events; and (3) a Granger causality attention mechanism that enhances prediction accuracy by emphasizing historically significant data patterns. Prophet surpasses State-Of-The-Art (SOTA) models like random forest, Long Short-Term Memory (LSTM), and other SOTA transformers in both accuracy and scalability.
This work determined the potential implication of firm specific factors on profitability of listed Insurance companies in Nigeria, 2015 to 2019. Firm specific factors are Firm Size, FSSIZE; Firm Age FSAGE; Firm Leverage FSLE; and Firm Liquidity FSLQ and the dependent variable is Profitability-Return on Asset ROA. This study applied ex-post facto research design and the population comprised all the quoted insurance companies in the Nigerian Stock Exchange NSE, 2014-2019. A purposive sampling technique selected 10 listed insurance firms with the required annual reports made available to the public in, 2020 NSE. The statistical techniques employed: Descriptive statistics; Pearson Correlation Matrix and Robust Least Square (RLS) Regression. The results show that the R-squared value is 0.0342, which implies that all the independent variables jointly explain only about 3.42% of the systematic variations in the (ROA). The final findings indicated that: FSSIZE has insignificant negative implication; FSAGE is positively insignificant; FSLEV is negatively insignificant; and FSLIQ are positively insignificant on, profitability of insurance firms pooled for the period. Our recommendations are that Insurance firms should increase firm size; earn more premiums to increase liquidity and leverage in order to increase profitability. We contribute with the findings that depict the true state of polled insurance companies in Nigeria, the modernized model and the rich literatures for academia. Implications are that insurance firms in increasing the firm size, and premium to drive profitability, and that Nigeria law only mandated third party vehicle insurance and also Nigerians' hostile attitude toward other insurance cover.
How does withdrawing from a free trade agreement affect maritime cargo volumes? Are there external effects on freight transport between third‐party countries? Using Brexit as a case study, in a difference‐in‐differences analysis of 2013–2024 port‐level data, I find a 21% decrease in EU‐UK roll‐on roll‐off (Ro‐Ro) volumes, and a 60% decrease in Ireland‐UK volumes. I find an 88% increase in Ireland‐France cargo, indicating a diversion from the UK ‘land‐bridge’ between Ireland and mainland Europe to the direct route. I estimate that emissions would be roughly 35% lower on the direct route, revealing an externality on transport between two third‐party countries.
In response to the onset of any black swan event i.e., COVID-19, the governments of global economies were prompted to implement stringent policies, overwhelming the extent and intensity of past pandemics e.g. Spanish Flu and the 1957-58 influenza outbreak to name the few. This study is an attempt to analyze the impact of the government stringent policies on oil price, exchange rates, stick returns, and commodity prices in order to combat the mortality rate of COVID-19. Our results illustrate that the commodity prices leave more definite impressions as compared to oil prices and exchange rates, while the stock returns are significantly impacted by government stringent measures. This study further accentuates the crucial role of policy actions for combating the economic effects of black swan event while backing the balanced measures which safeguard public health along with the economic stability.
This study investigates the role of financial development in environmental sustainability in the case of Nigeria using the recently developed Fourier ARDL (FARDL) and nonlinear ARDL (NARDL) estimation methods, for the first time. The FARDL model results reveal that financial development reduces carbon emissions in Nigeria in the long run, and the NARDL model validates the robustness of this finding. Based on our empirical findings, we recommend that the Nigerian government should promote the development of the financial sector by adopting an expansionary fiscal policy to boost incomes, investment, and economic growth.
Background: With COVID-19 having a significant impact on economic activity, it has become difficult for the existing dynamic factor models (nowcasting models) to forecast macroeconomics with high accuracy. The real-time monitoring of macroeconomics has become an important research problem faced by banks, governments, and corporations. Subjects and Methods: This paper proposes an adaptive evolutionary causal dynamic factor model (AcNowcasting) for macroeconomic forecasting. Unlike the classical nowcasting models, the AcNowcasting algorithm has the ability to perform feature selection. The criteria for feature selection are based on causality strength rather than being based on the quality of the prediction results. In addition, the factors in the AcNowcasting algorithm have the capacity for adaptive differential evolution, which can generate the best factors. These two abilities are not possessed by classical nowcasting models. Results: The experimental results show that the AcNowcasting algorithm can extract common factors that reflect macroeconomic fluctuations better, and the prediction accuracy of the AcNowcasting algorithm is more accurate than that of traditional nowcasting models. Contributions: The AcNowcasting algorithm provides a new prediction theory and a means for the real-time monitoring of macroeconomics, which has good theoretical and practical value.
Accurately predicting 5G bandwidth in vehicular mobility scenarios is essential for optimizing real-time communication (RTC) systems in a range of emerging vehicular applications, such as autonomous driving, in-car video conferencing, vehicular augmented reality (AR), and remote driving in dynamic urban environments. In this paper, we propose Prophet, a causality-aware Transformer model designed to forecast 5G throughput by capturing the complex causal relationships between control-plane 5G network events (e.g., handovers) and environmental factors (e.g., signal strength). Our key contributions include: (1) a wavelet-based encoder that efficiently captures long-term trends through multi-scale signal decomposition; (2) a control-plane causality attention mechanism in the decoder that focuses on localized, short-term throughput variations triggered by network events; and (3) a Granger causality attention mechanism that enhances prediction accuracy by emphasizing historically significant data patterns. Prophet surpasses State-Of-The-Art (SOTA) models like random forest, Long Short-Term Memory (LSTM), and other SOTA transformers in both accuracy and scalability.
Scholars of agenda setting have tried to untangle the relationship between the president and Congress. However, many studies treat Congress as if the House of Representatives and the Senate had one unified agenda. In this study, we examine the agenda‐setting relationship between the president and the House and Senate as separate institutions with their own agenda‐setting mechanisms to examine whether attention in one institution acts as a signal about policy problems for the other institutions. Examining the period of 1956 to 2017, we use Vector Autoregression to find the president dominates the agenda‐setting space, though the precise inter‐institutional agenda‐setting dynamics differ across issues and are not the same between the president and the House and Senate. Consequently, scholarly understandings of the congressional–presidential relationship must consider the independent relationships that presidents have with each chamber to understand the agenda‐setting power of the president.
Graph deep learning methods have become popular tools to process collections of correlated time series. Unlike traditional multivariate forecasting methods, graph-based predictors leverage pairwise relationships by conditioning forecasts on graphs spanning the time series collection. The conditioning takes the form of architectural inductive biases on the forecasting architecture, resulting in a family of models called spatiotemporal graph neural networks. These biases allow for training global forecasting models on large collections of time series while localizing predictions w.r.t. each element in the set (nodes) by accounting for correlations among them (edges). Recent advances in graph neural networks and deep learning for time series forecasting make the adoption of such processing framework appealing and timely. However, most studies focus on refining existing architectures by exploiting modern deep-learning practices. Conversely, foundational and methodological aspects have not been subject to systematic investigation. To fill this void, this tutorial paper aims to introduce a comprehensive methodological framework formalizing the forecasting problem and providing design principles for graph-based predictors, as well as methods to assess their performance. In addition, together with an overview of the field, we provide design guidelines and best practices, as well as an in-depth discussion of open challenges and future directions.
The continued rise in global carbon dioxide ([Formula: see text]) emissions challenges international climate policy, particularly the goals of the Paris Agreement. This study forecasts [Formula: see text] emissions through 2030 for the eleven highest-emitting nations-China, the United States, India, Russia, Japan, Iran, Indonesia, Saudi Arabia, Canada, South Korea, and Germany-while assessing their progress toward Nationally Determined Contributions (NDCs). Using data from 1990 to 2023, we apply a robust data pipeline comprised of six machine learning models and sequential squeeze feature selection incorporating eleven economic, industrial, and energy consumption variables. We have modelled the scenario with an average prediction accuracy of 96.21%. Results indicate that Russia is on track to exceed its reduction targets, while Germany and the United States will fall slightly short. China, India, Japan, Canada, South Korea, and Indonesia are projected to miss their commitments by significant margins. At the same time, Iran and Saudi Arabia are expected to increase emissions rather than reduce them. These findings highlight the need for strengthened energy efficiency policies, expanded renewable energy adoption, enhanced carbon pricing mechanisms, and stricter regulatory enforcement. Emerging economies require international collaboration and investment to support low-carbon transitions. This study provides a data-driven assessment of emission trajectories, emphasizing the urgency of coordinated global action, technological innovation, and adaptive policy measures to align emissions with the 1.5[Formula: see text] warming threshold. This work represents a novel integration of multivariate machine learning modelling, data-driven feature selection, and policy-oriented emission forecasts, establishing new methodological and empirical benchmarks in climate analytics.
Inferring time-varying gene regulatory networks from time-series single-cell RNA sequencing (scRNA-seq) data remains a challenging task. The existing methods have notable limitations as most are either designed for reconstructing time-varying networks from bulk microarray data or constrained to inferring stationary networks from scRNA-seq data, failing to capture the dynamic regulatory changes at the single-cell level. Furthermore, scRNA-seq data present unique challenges, including sparsity, dropout events, and the need to account for heterogeneity across individual cells. These challenges complicate the accurate capture of gene regulatory network dynamics over time. In this work, we propose a novel f-divergence-based dynamic gene regulatory network inference method (f-DyGRN), which applies f-divergence to quantify the temporal variations in gene expression across individual single cells. Our approach integrates a first-order Granger causality model with various regularization techniques and partial correlation analysis to reconstruct gene regulatory networks from scRNA-seq data. To infer dynamic regulatory networks at different stages, we employ a moving window strategy, which allows for the capture of dynamic changes in gene interactions over time. We applied this method to analyze both simulated and real scRNA-seq data from THP-1 human myeloid monocytic leukemia cells, comparing its performance with the existing approaches. Our results demonstrate that f-DyGRN, when equipped with a suitable f-divergence measure, outperforms most of the existing methods in reconstructing dynamic regulatory networks from time-series scRNA-seq data.
This paper, which in part serves as a common introduction to the two papers following in this issue, attempts to define the meaning of the "casual interpretability of a parameter" in a system of simultaneous linear relationships. It attempts, moreover, to expound a basis for interpreting the parameters of a nonrecursive or interdependent system casually. This is done in terms of an underlying causal chain system to which the interdependent system is either an approximation or a description of the equilibrium state.
This paper discusses one of the uses to which two powerful techniques of modern time series analysis may be put in economics: namely, the study of the precise effects of seasonal adjustment procedures on the characteristics of the series to which they are applied. Since most economic data appearing at intervals of less than a year are to a greater or lesser extent "manufactured" from more basic time series, the problem of assessing the effects of the "manufacturing" processes upon the essential characteristics of the raw material to which they are applied is not unimportant. Perhaps the most common type of adjustment applied to raw economic time series is that designed to eliminate so-called seasonal fluctuations. The precise nature of seasonality is not easy to define, but an attempt is made in Section 2.1 below. The techniques employed to study the effects of seasonal adjustment procedures are those of spectral and cross-spectral analysis. In somewhat oversimplified terms the basic idea behind these types of analysis is that a stochastic time series may be decomposed into an infinite number of sine and cosine waves with infinitesimal random amplitudes. Spectral analysis deals with a single time series in terms of its frequency "content": cross-spectral analysis deals with the relation between two time series in terms of their respective frequency "contents." The two techniques are discussed in both theoretical and practical terms. Spectral analyses have been made for about seventy-five time series of United States employment, unemployment, labor force, and various categories thereof. Cross-spectral analyses have been made of the relations between these series and the corresponding series as seasonally adjustment by the procedures used by the Bureau of Labor Statistics. Two major conclusions regarding the effects of the BLS seasonal adjustment procedures emerge from these emerge from these analyses. First, these procedures remove far more from the series to which they are applied than can properly be considered as seasonal. Second, if the relation between two seasonally adjusted series in time is compared with the corresponding relation between the original series in time, it is found that there is a distortion due to the process of seasonal adjustment itself. Both defects impair the usefulness of the seasonally adjusted series as indicators of economic conditions, but, of the two, temporal distortion is the more serious defect. Examples of some of these results are discussed below in Section 3.3.