arXiv · 2611.20009
Testing Cross-Sectional Dependence in Clustered Data
Abstract
In clustered data, cross-sectional dependence within clusters can be characterized as weak, semi-strong, or strong. This paper proposes a simple test to identify the specific nature of this dependence in linear regression models. By exploiting the variation in cluster sizes, our approach detects the underlying dependence structure in a computationally simple and robust manner. We develop a test statistic that is shown to be asymptotically chi-square distributed under the null hypothesis of weak dependence. Simulation studies and an empirical illustration using India's Household Consumption Expenditure Survey (HCES) data for 2022-23 demonstrate the usefulness of the proposed test.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Subhodeep Dey, Gopal K. Basak, Samarjit Das. 2026-11-15. Testing Cross-Sectional Dependence in Clustered Data. https://searcharxiv.com/abs/2611.20009
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.