EconPapers    
Economics at your fingertips  
 

Cluster Sample Inference with Very Few Groups

Johan Vikström
Additional contact information
Johan Vikström: IFAU Uppsala, Box 513, S-752 20 Uppsala, Sweden and UCLS, Department of Economics University, johan.vikstrom@ifau.uu.se

Econometrics Letters, 2016, vol. 3, issue 1, 1-10

Abstract: If the number of clusters equals the number of cluster-invariant regressors the usual clusters-specific standard errors cannot be computed. One important example is the difference in differences setting with two groups and two time periods. This paper introduces a new method to calculate how large the within cluster correlation of the model errors can be yet still enable OLS coefficients to retain statistical significance at a desired level. This sensitivity approach is applied to previous results regarding the effects of workers compensation benefits.

Keywords: Cluster effect; cluster robust; difference in difference; sensitivity; workers compensation. (search for similar items in EconPapers)
JEL-codes: C12 C21 C23 (search for similar items in EconPapers)
Date: 2016
References: Add references at CitEc
Citations Track citations by RSS feed

Downloads: (external link)
http://dx.doi.org/10.5455/Elet.2016.3.1.1

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:bmo:bmoart:v:3:y:2016:i:1:p:1-10

Access Statistics for this article

More articles in Econometrics Letters from Bilimsel Mektuplar Organizasyonu (Scientific letters)
Series data maintained by Tolga Omay ().

 
Page updated 2017-09-29
Handle: RePEc:bmo:bmoart:v:3:y:2016:i:1:p:1-10