The bootstrap is a method for estimating the distribution of an estimator or test statistic by resampling one’s data. It is often much more accurate in finite samples than ordinary asymptotic approximations are. This is important in applied research, because the familiar asymptotic normal and chi-square approximations can be very inaccurate. When this happens, the difference between the true and nominal coverage probability of a confidence interval or rejection probability of a test can be very large, and inference can be highly misleading. The bootstrap often greatly reduces errors in coverage and rejection probabilities, thereby making reliable inference possible.
KeywordsAsymptotic distribution Asymptotic refinements Bias reduction Bootstrap Conditional Kolmogorov test statistic Edgeworth approximations Maximum score estimator Monte Carlo simulation Probability Probit models Statistical inference Statistics and economics Subsampling Tobit model
I thank Federico Bugni for helpful comments. The preparation of this article was supported in part by NSF Grant SES-0352675.
- Beran, R., and G.R. Ducharme. 1991. Asymptotic theory for bootstrap methods in statistics. Montréal: Les Publications CRM, Centre de Recherches Mathematiques, Université de Montréal.Google Scholar
- Hall, P. 1994. Methodology and theory for the bootstrap. In Handbook of econometrics, ed. R.F. Engle and D.F. McFadden, Vol. 4. Amsterdam: North-Holland.Google Scholar
- Horowitz, J.L. 1997. Bootstrap methods in econometrics: Theory and numerical performance. In Advances in economics and econometrics: Theory and applications, seventh world congress, ed. D.M. Kreps and K.F. Wallis, Vol. 3. Cambridge: Cambridge University Press.Google Scholar
- Horowitz, J.L. 2001. The bootstrap in econometrics. In Handbook of econometrics, ed. J.J. Heckman and E.E. Leamer, Vol. 5. Amsterdam: North-Holland.Google Scholar
- Maddala, G.S., and J. Jeong. 1993. A perspective on application of bootstrap methods in econometrics. In Handbook of statistics, ed. G.S. Maddala, C.R. Rao, and H.D. Vinod, Vol. 11. Amsterdam: North-Holland.Google Scholar
- Vinod, H.D. 1993. Bootstrap methods: Applications in econometrics. In Handbook of statistics, ed. G.S. Maddala, C.R. Rao, and H.D. Vinod, Vol. 11. Amsterdam: North-Holland.Google Scholar