M
Matthew D. Webb
Researcher at Carleton University
Publications - 35
Citations - 1502
Matthew D. Webb is an academic researcher from Carleton University. The author has contributed to research in topics: Inference & Estimator. The author has an hindex of 10, co-authored 31 publications receiving 954 citations. Previous affiliations of Matthew D. Webb include University of Calgary.
Papers
More filters
Journal ArticleDOI
Fast and wild: Bootstrap inference in Stata using boottest:
TL;DR: The wild bootstrap was originally developed for regression models with heteroskedasticity of unknown form and has been extended to models estimated by instrumental variables over the past 30 years as discussed by the authors.
Journal ArticleDOI
Wild Bootstrap Inference for Wildly Different Cluster Sizes
TL;DR: In this article, the authors show that the rule of 42 is not true for unbalanced clusters and use critical values based on the wild cluster bootstrap to improve the performance of CRVE.
Posted ContentDOI
Reworking Wild Bootstrap Based Inference for Clustered Errors
TL;DR: It is demonstrated, using Monte Carlo experiments, that a 6-point bootstrap weight distribution improves the reliability of inference with few clusters, and two alternative wild bootstrap procedures are suggested.
Journal ArticleDOI
The wild bootstrap for few (treated) clusters
TL;DR: In this paper, a family of subcluster wild bootstraps for linear regression models is proposed, which includes the ordinary wild bootstrap as a limiting case for pure treatment models.
Posted Content
Fast And Wild: Bootstrap Inference In Stata Using Boottest
TL;DR: The main ideas of the wild cluster bootstrap are reviewed, tips for use are offered, why it is particularly amenable to computational optimization is explained, and the syntax of boottest, artest, scoretest, and waldtest is state.