Direction Dependence Analysis

Erscheinungsjahr 2025
ISBN: 978-1-009-38141-3
Verlag: Cambridge University Press

Buch, Englisch, 388 Seiten, Format (B × H): 157 mm x 235 mm, Gewicht: 711 g

While regression analysis is widely understood, it falls short in determining the causal direction of relationships in observational data. In this groundbreaking volume, Wiedermann and von Eye introduce Direction Dependence Analysis (DDA), a novel method that leverages variable information often overlooked by traditional techniques, such as higher-order moments like skewness and kurtosis. DDA reveals the asymmetry properties of regression and correlation, enabling researchers to evaluate competing causal hypotheses, assess the roles of variables in causal flows, and develop statistical methods for testing causal direction. This book provides a comprehensive formal description of DDA, illustrated with both artificial and real-world data examples. Additionally, readers will find free software implementations of DDA, making this an essential resource for researchers seeking to enhance their understanding of causal relationships in data analysis.

Autoren/Hrsg.

Wiedermann, Wolfgang

Eye, Alexander Von

Inhaltsverzeichnis

1. Introduction; 2. The linear regression model; 3. Asymmetry properties of distributions of observed variables; 4. Asymmetry properties of error distributions; 5. Independence properties of causes and errors; 6. Direction of dependence under latent confounding; 7. The integrated framework of Direction Dependence Analysis; 8. Stability and sensitivity analyses; 9. Extensions and applications; 10. Statistical software; 11. Concluding remarks.

Über Autor(innen)

Wiedermann, Wolfgang
Wolfgang Wiedermann is Professor of Statistics, Measurement, and Evaluation in Education in the College of Education and Human Development at the University of Missouri, Columbia. He received his Ph.D. in Quantitative Psychology from the University of Klagenfurt, Austria. His work focuses on the development of methods for causal structure learning and causal inference, distributional regression, and methods for person-oriented research. He has co-authored books on the general linear model (in 2023) and Configural Frequency Analysis (in 2021) and edited volumes on direction dependence modeling (in 2020) and statistics and causality (in 2016). His work appears in journals such as Psychological Methods, Multivariate Behavioral Research, Behavior Research Methods, Prevention Science, Developmental Psychology, and Development and Psychopathology.

Eye, Alexander Von
Alexander von Eye, is Professor Emeritus of Psychology at Michigan State University. He received his Ph.D. in Psychology from the University of Trier, Germany, in 1976. His work focuses on categorical data analysis, methods of analysis of direction dependence hypotheses, person-oriented research, and human development. He authored texts on, e.g., Configural Frequency Analysis (with Wiedermann), and on log-linear modeling, and he edited, e.g., a book on statistics and causality (with Wiedermann). His over 400 articles appeared in the premier journals of the field, including Psychological Methods, Multivariate Behavioral Research, Child Development, the American Statistician, and the Journal of Applied Statistics.