Exploring the Relationship Between Temporal Correlation and Variance Inflation in Time Series Regression Models
Posted: Nov 16, 2019
Abstract
The analysis of time series data represents a cornerstone of empirical research across numerous disciplines including economics, finance, environmental science, and engineering. Traditional time series regression methodologies have evolved substantially since the pioneering work of Box and Jenkins, yet certain fundamental assumptions regarding the independence of various statistical phenomena persist in both theoretical development and practical application. Specifically, the relationship between temporal correlation structures and multicollinearity-induced variance inflation has received remarkably limited attention in the extant literature, despite their frequent co-occurrence in empirical applications. This research gap is particularly concerning given that both phenomena independently represent significant threats to the validity of statistical inference in regression contexts. This paper challenges the compartmentalized approach by demonstrating that temporal correlation and variance inflation interact in complex ways that are not adequately addressed by current methodologies.
Downloads: 94
Abstract Views: 1352
Rank: 318457