Econometrica: Sep 2021, Volume 89, Issue 5

Bootstrap with Cluster-Dependence in Two or More Dimensions

https://doi.org/10.3982/ECTA15383
p. 2143-2188

Konrad Menzel

We propose a bootstrap procedure for data that may exhibit cluster‐dependence in two or more dimensions. The asymptotic distribution of the sample mean or other statistics may be non‐Gaussian if observations are dependent but uncorrelated within clusters. We show that there exists no procedure for estimating the limiting distribution of the sample mean under two‐way clustering that achieves uniform consistency. However, we propose bootstrap procedures that achieve adaptivity with respect to different uniformity criteria. Important cases and extensions discussed in the paper include regression inference, U‐ and V‐statistics, subgraph counts for network data, and non‐exhaustive samples of matched data.



Log In To View Full Content

Supplemental Material

Supplement to "Bootstrap with Cluster-Dependence in Two or More Dimensions"

This zip file contains the replication files for the manuscript.

Read More View ZIP


Supplement to "Bootstrap with Cluster-Dependence in Two or More Dimensions"

This appendix contains extensions of the results in Menzel (2021) as well as an asymptotic theory for alternative inference procedures under multi-way clustering.

Read More View PDF



Back