Bootstrap-Based Improvements for Inference with Clustered Errors. Cameron, A. C., Gelbach, J. B., & Miller, D. L. September, 2007. ![Bootstrap-Based Improvements for Inference with Clustered Errors [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex Researchers have increasingly realized the need to account for within-group dependence in estimating standard errors of regression parameter estimates. The usual solution is to calculate cluster-robust standard errors that permit heteroskedasticity and within-cluster error correlation, but presume that the number of clusters is large. Standard asymptotic tests can over-reject, however, with few (5-30) clusters. We investigate inference using cluster bootstrap-t procedures that provide asymptotic refinement. These procedures are evaluated using Monte Carlos, including the example of Bertrand, Duflo and Mullainathan (2004). Rejection rates of ten percent using standard methods can be reduced to the nominal size of five percent using our methods.
@misc{cameron_bootstrap-based_2007,
type = {Working {Paper}},
series = {Technical {Working} {Paper} {Series}},
title = {Bootstrap-{Based} {Improvements} for {Inference} with {Clustered} {Errors}},
url = {https://www.nber.org/papers/t0344},
doi = {10.3386/t0344},
abstract = {Researchers have increasingly realized the need to account for within-group dependence in estimating standard errors of regression parameter estimates. The usual solution is to calculate cluster-robust standard errors that permit heteroskedasticity and within-cluster error correlation, but presume that the number of clusters is large. Standard asymptotic tests can over-reject, however, with few (5-30) clusters. We investigate inference using cluster bootstrap-t procedures that provide asymptotic refinement. These procedures are evaluated using Monte Carlos, including the example of Bertrand, Duflo and Mullainathan (2004). Rejection rates of ten percent using standard methods can be reduced to the nominal size of five percent using our methods.},
urldate = {2023-12-12},
publisher = {National Bureau of Economic Research},
author = {Cameron, A. Colin and Gelbach, Jonah B. and Miller, Douglas L.},
month = sep,
year = {2007},
doi = {10.3386/t0344},
}
Downloads: 0
{"_id":"wwEJodk9fLmhgcFGZ","bibbaseid":"cameron-gelbach-miller-bootstrapbasedimprovementsforinferencewithclusterederrors-2007","author_short":["Cameron, A. C.","Gelbach, J. B.","Miller, D. L."],"bibdata":{"bibtype":"misc","type":"Working Paper","series":"Technical Working Paper Series","title":"Bootstrap-Based Improvements for Inference with Clustered Errors","url":"https://www.nber.org/papers/t0344","doi":"10.3386/t0344","abstract":"Researchers have increasingly realized the need to account for within-group dependence in estimating standard errors of regression parameter estimates. The usual solution is to calculate cluster-robust standard errors that permit heteroskedasticity and within-cluster error correlation, but presume that the number of clusters is large. Standard asymptotic tests can over-reject, however, with few (5-30) clusters. We investigate inference using cluster bootstrap-t procedures that provide asymptotic refinement. These procedures are evaluated using Monte Carlos, including the example of Bertrand, Duflo and Mullainathan (2004). Rejection rates of ten percent using standard methods can be reduced to the nominal size of five percent using our methods.","urldate":"2023-12-12","publisher":"National Bureau of Economic Research","author":[{"propositions":[],"lastnames":["Cameron"],"firstnames":["A.","Colin"],"suffixes":[]},{"propositions":[],"lastnames":["Gelbach"],"firstnames":["Jonah","B."],"suffixes":[]},{"propositions":[],"lastnames":["Miller"],"firstnames":["Douglas","L."],"suffixes":[]}],"month":"September","year":"2007","bibtex":"@misc{cameron_bootstrap-based_2007,\n\ttype = {Working {Paper}},\n\tseries = {Technical {Working} {Paper} {Series}},\n\ttitle = {Bootstrap-{Based} {Improvements} for {Inference} with {Clustered} {Errors}},\n\turl = {https://www.nber.org/papers/t0344},\n\tdoi = {10.3386/t0344},\n\tabstract = {Researchers have increasingly realized the need to account for within-group dependence in estimating standard errors of regression parameter estimates. The usual solution is to calculate cluster-robust standard errors that permit heteroskedasticity and within-cluster error correlation, but presume that the number of clusters is large. Standard asymptotic tests can over-reject, however, with few (5-30) clusters. We investigate inference using cluster bootstrap-t procedures that provide asymptotic refinement. These procedures are evaluated using Monte Carlos, including the example of Bertrand, Duflo and Mullainathan (2004). Rejection rates of ten percent using standard methods can be reduced to the nominal size of five percent using our methods.},\n\turldate = {2023-12-12},\n\tpublisher = {National Bureau of Economic Research},\n\tauthor = {Cameron, A. Colin and Gelbach, Jonah B. and Miller, Douglas L.},\n\tmonth = sep,\n\tyear = {2007},\n\tdoi = {10.3386/t0344},\n}\n\n\n\n\n\n\n\n\n\n\n\n","author_short":["Cameron, A. C.","Gelbach, J. B.","Miller, D. L."],"key":"cameron_bootstrap-based_2007","id":"cameron_bootstrap-based_2007","bibbaseid":"cameron-gelbach-miller-bootstrapbasedimprovementsforinferencewithclusterederrors-2007","role":"author","urls":{"Paper":"https://www.nber.org/papers/t0344"},"metadata":{"authorlinks":{}},"html":""},"bibtype":"misc","biburl":"https://bibbase.org/zotero/studhall","dataSources":["P2gwcipDHM5rpi2bX"],"keywords":[],"search_terms":["bootstrap","based","improvements","inference","clustered","errors","cameron","gelbach","miller"],"title":"Bootstrap-Based Improvements for Inference with Clustered Errors","year":2007}