Review
Mathematics, Applied
Tom Edinburgh, Stephen J. Eglen, Ari Ercole
Summary: Inferring nonlinear and asymmetric causal relationships between multivariate longitudinal data is crucial in various fields such as clinical medicine, mathematical biology, economics, and environmental research. Evaluation of ten prominent causality indices showed strong agreement between methods in general, but they may not always be robust to real-world relevant transformations.
Article
Multidisciplinary Sciences
Manuel Castro, Pedro Ribeiro Mendes Junior, Aurea Soriano-Vargas, Rafael de Oliveira Werneck, Maiara Moreira Goncalves, Leopoldo Lusquino Filho, Renato Moura, Marcelo Zampieri, Oscar Linares, Vitor Ferreira, Alexandre Ferreira, Alessandra Davolio, Denis Schiozer, Anderson Rocha
Summary: In this study, we propose using ensemble models (such as Random Forest) to assess the importance of input features in machine learning models, in order to establish causal relationships between variables. By analyzing oil field production data, we find that our results align with confirmed tracer information, demonstrating the effectiveness of our proposed methodology.
SCIENTIFIC REPORTS
(2023)
Article
Physics, Multidisciplinary
Christos Koutlis, Dimitris Kugiumtzis
Summary: Various methods of Granger causality have been developed to analyze causal relationships between system variables. The impact of unobserved variables on connectivity structure estimation has been studied, with a focus on estimating direct causality effects in high-dimensional time series.
Article
Multidisciplinary Sciences
Axel Wismueller, Adora M. Dsouza, M. Ali Vosoughi, Anas Abidin
Summary: The study introduces a new method, lsNGC, that can identify causal relationships from limited observational data without explicit a priori assumptions of functional interdependence between component time series.
SCIENTIFIC REPORTS
(2021)
Article
Engineering, Aerospace
Luisina Pastorino, Massimiliano Zanin
Summary: Complex network theory, along with metrics for detecting causality relationships from time series, has been increasingly used to study delay propagation in air transport. To convert landing events into time series, most previous works have focused on fixed-size windows. However, we demonstrate that an optimal airport-specific window size can be calculated to maximize the number of detected causality relationships. We also show that this choice affects the macro-scale structure and airport centrality, highlighting its importance in delay propagation.
Article
Mathematics, Applied
Angeliki Papana, Elsa Siggiridou, Dimitris Kugiumtzis
Summary: The concept of Granger causality is increasingly used to characterize directional interactions, with a multivariate framework being essential for estimating Granger causality from multivariate time series. Direct causality measures with variable selection and dimension reduction techniques have been introduced to address estimation problems related to non-informative or non-significant variables. The study shows the superiority of dimension reduction measures, especially for high-dimensional systems, in assessing the performance of bivariate and multivariate causality measures in the time domain.
COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION
(2021)
Article
Computer Science, Information Systems
Ziheng Duan, Haoyan Xu, Yida Huang, Jie Feng, Yueyang Wang
Summary: In this paper, a novel end-to-end deep learning model called CauGNN is proposed to tackle the problem of multivariate time series (MTS) forecasting. By introducing the neural Granger causality graph to characterize the causal relationships among variables, using convolutional neural network filters for time series feature extraction, and employing graph neural network for forecasting the graph structure generated by MTS, the proposed method achieves state-of-the-art results in MTS forecasting task.
TSINGHUA SCIENCE AND TECHNOLOGY
(2023)
Article
Mathematics
Achilleas Anastasiou, Peter Hatzopoulos, Alex Karagrigoriou, George Mavridoglou
Summary: This work focuses on developing new distance measure algorithms for analyzing causal relationships in financial and economic data. The proposed methodology was applied to a case study involving the classification of 19 EU countries based on health resource variables.
Review
Biology
Alex Eric Yuan, Wenying Shou
Summary: This article provides a critical review of three statistical causal discovery methods and their applications in ecological processes. The review examines what each method tests for, the causal statements it implies, and the potential for misinterpretation. The authors introduce new visualization techniques and highlight the limitations of so-called "model-free" causality tests. The goal of the review is to encourage thoughtful application of these methods, facilitate interdisciplinary communication, and promote explicit assumptions.
Article
Economics
Han Lin Shang, Kaiying Ji, Ufuk Beyaztas
Summary: By studying causality between bivariate curve time series using Granger causality generalized measures of correlation, we can determine which curve time series influences the other and the predictability of the two series. An example in climatology shows that sea surface temperature Granger-causes sea-level atmospheric pressure, while in finance, we identify stocks that lead or lag behind Dow Jones industrial averages. The close relationship between the S&P 500 index and crude oil price helps us determine leading and lagging variables.
JOURNAL OF FORECASTING
(2021)
Review
Environmental Sciences
Jakob Runge, Andreas Gerhardus, Gherardo Varando, Veronika Eyring, Gustau Camps-Valls
Summary: Many research questions in Earth and environmental sciences require causal inference to establish relationships between variables. However, there is a language gap between methodological and domain science communities. This Technical Review explains the use of causal inference in time series data and provides practical case studies to address challenges and improve data-driven science in Earth and environmental sciences.
NATURE REVIEWS EARTH & ENVIRONMENT
(2023)
Article
Physics, Multidisciplinary
Martina Chvostekova, Jozef Jakubik, Anna Krakovska
Summary: The study investigates the information flow time arrow for stochastic data defined by vector autoregressive models. It analyzes time series forward and backward using different Granger causality detection methods, considering distributions of predictive errors other than the normal distribution. The study finds a clear effect of a change in the order of cause and effect on time-reversed series of unidirectionally connected variables under certain conditions.
Article
Automation & Control Systems
Jince Li, Yilin Shi, Hongguang Li, Bo Yang
Summary: In this article, a novel prediction model called temporal causal graph attention networks with nonlinear paradigms (TC-GATN) is proposed to capture inherent dependencies on industrial multivariate time-series (MTS). The model utilizes a graph learning algorithm based on Granger causality to extract potential relationships among multiple variables and guide the directional edge connections of the hierarchy. Parallel gated recurrent unit encoders are introduced to perform the nonlinear interaction of node features, while the self-attention mechanism aggregates encoder hidden states across all stages. A temporal module is added to process information from the graph layer, achieving satisfactory predictions. The feasibility and effectiveness of TC-GATN are validated using two actual datasets.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Mathematics, Interdisciplinary Applications
Akylas Fotiadis, Dimitris Kugiumtzis
Summary: Estimating the interdependence structure of complex dynamical systems is challenging, especially with the availability of more observed variables in the era of Big Data. Dimension reduction is necessary in Granger causality measure to estimate direct causality effects, while also detecting non-linear effects. PMIME, a model-free information-based measure, performs well on moderately high-dimensional multivariate time series. This study investigates the formation of complex networks from high-dimensional time series and explores the effect of network size using PMIME on coupled dynamical systems and the British stock market during Brexit.
JOURNAL OF COMPLEX NETWORKS
(2023)
Article
Social Sciences, Interdisciplinary
Hao Yang, Shaobin Wang, Zhoupeng Ren, Haimeng Liu, Yun Tong, Na Wang
Summary: This paper investigates the dynamic relationship between life expectancy and various influencing factors in Beijing City. The study finds that environmental factors, such as air pollution and green land area, have a stronger impact on life expectancy than socioeconomic factors in Beijing.
SOCIAL INDICATORS RESEARCH
(2022)
Article
Green & Sustainable Science & Technology
Manoela Cabo, Elton Fernandes, Paulo Alonso, Ricardo Rodrigues Pacheco, Felipe Fagundes
Article
Engineering, Industrial
Flavio Soares de Oliveira Junior, Elton Fernandes, Laura Bahiense, Carlos Moacir Grandi
Summary: This research aims to develop and test a mathematical approach to estimate the optimal aircraft age for decision-making regarding withdrawal, market relocation, or retirement and decommissioning. The core contribution lies in the systematic analysis of opportunities to return parked aircraft to active service, considering scenarios of revenue improvements or cost reductions, to postpone retirement decisions. The study focuses on financial concerns of end-of-life aircraft, in estimating parking and retirement costs, revenue recovery from market relocation, and recovered values from disassembly and dismantling, for well-supported fleet planning.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2021)
Article
Green & Sustainable Science & Technology
Rodrigo V. Ventura, Manoela Cabo, Rafael Caixeta, Elton Fernandes, Vicente Aprigliano Fernandes
Article
Green & Sustainable Science & Technology
Vicente Aprigliano Fernandes, Ricardo R. Pacheco, Elton Fernandes, Manoela Cabo, Rodrigo V. Ventura, Rafael Caixeta
Summary: The study investigates the causal relationships between regular domestic air passenger transport links and the gross domestic product of small municipalities in remote areas, finding that the gross domestic product has a stronger impact on air transport than vice versa.
Article
Transportation
Rodrigo V. Ventura, Elton Fernandes, Vicente Aprigliano Fernandes, Manoela Cabo, Augusto Cesar Fadel, Rafael Caixeta
Summary: This paper investigates the income-price elasticities of passenger demand for flight connections in Brazil, considering the diversity of different regions and time periods. The results show variations in elasticities based on geographical location and macroeconomic conditions, shedding light on the demand elasticity dynamics and its strategic implications for the Brazilian civil aviation sector.
CASE STUDIES ON TRANSPORT POLICY
(2022)
Article
Green & Sustainable Science & Technology
Vicente Aprigliano Fernandes, Ricardo Rodrigues Pacheco, Elton Fernandes, Manoela Cabo, Rodrigo V. Ventura
Summary: This study aims to understand the development of Brazil's domestic air passenger network by analyzing passenger origins and destinations. The findings indicate that emerging origins and destinations, such as those connecting the Northeast region, show greater strength and potential. The analysis of specific links also reveals that important Brazilian airports may not necessarily have higher competence in generating air travel.
Article
Transportation
Vicente Aprigliano Fernandes, Ricardo Rodrigues Pacheco, Elton Fernandes
Summary: This paper analyzes the relationship between air transport and tourism in Brazil and finds that the efficiency of Brazil's tourism industry has deteriorated. It suggests the need to focus on regional development, particularly in the North and Northeast regions.
JOURNAL OF AIR TRANSPORT MANAGEMENT
(2022)
Article
Hospitality, Leisure, Sport & Tourism
Vicente Aprigliano Fernandes, Ricardo Rodrigues Pacheco, Elton Fernandes
Summary: This study examines the impact of air transport on tourism-related employment in Brazil, finding that air transport activity significantly affects employment in the tourism industry, with domestic air transport considered the main driver of employment. Brazilian policymakers must consider domestic air transport as a key factor for sustainable employment in the tourism industry.
CURRENT ISSUES IN TOURISM
(2021)
Article
Economics
Vicente Aprigliano Fernandes, Ricardo Rodrigues Pacheco, Elton Fernandes, William Ribeiro da Silva
JOURNAL OF TRANSPORT GEOGRAPHY
(2019)
Article
Hospitality, Leisure, Sport & Tourism
Elton Fernandes, Ricardo Rodrigues Pacheco, Vicente Aprigliano Fernandes
INTERNATIONAL JOURNAL OF TOURISM RESEARCH
(2019)
