Bootstrap-Based Improvements for Inference with Clustered Errors

By Colin Cameron, Jonah B. Gelbach, Douglas L. Miller, U. C.-Davis. Review of Economics and Statistics, 2008

3,130 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/f74ab0504b5f9618dadc68f40c2a796febefe74c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

The Rise and Decline of General Laws of Capitalism (2015, positive): “Standard errors allowing for arbitrary heteroscedasticity and serial correlation of residuals at the country level are computed using the pairs-cluster bootstrap procedure proposed by Cameron, Gelbach and Miller (2008) and are reported in parentheses.”
Inference in Differences-in-Differences with Few Treated Groups and Heteroskedasticity (2019, positive): “In this case, the bootstrap distribution will not be smooth even with many bootstrap replications (Cameron et al., 2008).”
The Institutional Causes of China's Great Famine, 1959–1961 (2015, positive): “Our results are similar if we estimate wild bootstrapped standard errors that are clustered at the province level and adjust for the small number of clusters (Cameron et al. 2008).”
Schooling, Violent Conflict, and Gender in Burundi (2011, positive): “We therefore use the bootstrapping method proposed by Cameron et al. (2008) to improve the estimation.”
Robust Inference With Multiway Clustering (2006, neutral): “This bootstrap provides the same level of asymptotic approximation as that without bootstrap, and does not additionally provide an asymptotic renement (see Cameron et al. (2008) for a discussion of clustering and asymptotic renement in the one-way case).”
The effect of aid on growth: evidence from a Quasi-experiment (2014, neutral): “Although 35 clusters is not a small number, for robustness we also report the standard errors from the clustered wild bootstrap procedure following Cameron et al. (2008). recommends the clustered wild bootstrap-t procedure for better asymptotic properties, as the t statistic is “pivotal”.”
Small-Sample Methods for Cluster-Robust Variance Estimation and Hypothesis Testing in Fixed Effects Models (2016, positive): “alar parameters that maintains the nominal Type-I error rate; however, their method requires that the target parameter be identied within each independent cluster and so it is not always applicable. Cameron et al. (2008) investigated a range of bootstrapping 5 procedures that provide improved Type-I”
How did distributional preferences change during the Great Recession (2015, positive): “we report OLS specifications clustered by session and generate p-values using the Wild cluster bootstrapping procedure described in Cameron, Gelbach and Miller (2008) and Cameron and Miller (forthcoming) to correct for the small number of sessions.”
What can we learn from benefit transfer errors? Evidence from 20 years of research on convergent validity (2013, neutral): “[7] demonstrate that these standard errors perform well in making robust inferences in the presence of clustered observations.”