Davidson, Russell; MacKinnon, James G. Wild bootstrap tests for IV regression. (English) Zbl 1198.62035 J. Bus. Econ. Stat. 28, No. 1, 128-144 (2010). Summary: We propose a wild bootstrap procedure for linear regression models estimated by instrumental variables. Like other bootstrap procedures that we proposed elsewhere, it uses efficient estimates of the reduced-form equation\((s)\). Unlike earlier procedures, it takes account of possible heteroscedasticity of unknown form. We apply this procedure to \(t\) tests, including heteroscedasticity-robust \(t\) tests, and to the T. W. Anderson and H. Rubin [Ann. Math. Stat. 20, 46–63 (1949; Zbl 0033.08002)] test. We provide simulation evidence that it works far better than older methods, such as the pairs bootstrap. We also show how to obtain reliable confidence intervals by inverting bootstrap tests. An empirical example illustrates the utility of these procedures. Cited in 1 ReviewCited in 10 Documents MSC: 62G09 Nonparametric statistical resampling methods 62J05 Linear regression; mixed models 65C60 Computational problems in statistics (MSC2010) 62G15 Nonparametric tolerance and confidence regions 62G10 Nonparametric hypothesis testing Keywords:Anderson-Rubin test; confidence interval; instrumental variables estimation; pairs bootstrap; residual bootstrap; two-stage least squares; weak instruments; wild bootstrap Citations:Zbl 0033.08002 PDFBibTeX XMLCite \textit{R. Davidson} and \textit{J. G. MacKinnon}, J. Bus. Econ. Stat. 28, No. 1, 128--144 (2010; Zbl 1198.62035) Full Text: DOI Link