Template-Type: ReDIF-Paper 1.0
Author-Name: Jonah B. Gelbach
Author-Name-First: Jonah B.
Author-Name-Last: Gelbach
Author-Name: Doug Miller
Author-Name-First: Doug
Author-Name-Last: Miller
Author-Name: A. Colin Cameron
Author-Name-First: A. Colin
Author-Name-Last: Cameron
Author-Workplace-Name: Department of Economics, University of California Davis
Title: Bootstrap-Based Improvements for Inference with Clustered Errors
Abstract: Microeconometrics researchers have increasingly realized the essential need to account for any within-group dependence in estimating standard errors of regression parameter estimates. The typical preferred solution is to calculate cluster-robust or sandwich standard errors that permit quite general heteroskedasticity and within-cluster error correlation, but presume that the number of clusters is large. In applications with few (5-30) clusters, standard asymptotic tests can over-reject considerably. We investigate more accurate inference using cluster bootstrap-t procedures that provide asymptotic refinement. These procedures are evaluated using Monte Carlos, including the much-cited differences-in-differences example of Bertrand, Mullainathan and Duflo (2004). In situations where standard methods lead to rejection rates in excess of ten percent (ormore) for tests of nominal size 0.05, our methods can reduce this to five percent. In principle a pairs cluster bootstrap should work well, but in practice a Wild cluster bootstrap performs better.
Length: 52
File-URL: https://repec.dss.ucdavis.edu/files/QpaWYKbXUsANCptT7guxmWaK/06-21.pdf
File-Format: application/pdf
Number: 128
Classification-JEL: C15, C12, C21
KeyWords: clustered errors; random effects; cluster robust; sandwich; bootstrap; bootstrap-t; clustered bootstrap; pairs bootstrap; wild bootstrap.
Creation-Date: 20060713
Handle: RePEc:cda:wpaper:128