On bootstrapping regression model with random resample size
Corressponding author's email:
nhungnh@hcmute.edu.vnKeywords:
bootstrap, regression, resampling, random resample size, uniform distributionAbstract
To find confidence interval for regression coefficients, classical methods require the distribution of errors. Under mild conditions, without knowing the distribution of errors, the bootstrap approximation with fixed or random resample sizeto estimate the distribution of the least squares is valid. In this paper, the author presents algorithms to determine regression coefficients of the bootstrap regression model with random resample size Nn.Nn is a positive integer-valued in m, n with the ability to be the same at all values, where m is the smallest positive integer greater than or equal to n/4. Matlab software is used to seek the empirical bootstrap regression coefficients and create analysis comments.
Downloads: 0
References
Bradley Efron. Bootstrap method: Another look at the Jacknife. Ann. Statist. 7. (1979).
E. Mammen. Bootstrap, wild bootstrap, and asymptotic normality. Probab. Theory Relat. Fields 93, 439–455 (1992).
C. R. Rao, P.K. Pathak, and V. I. Koltchinskii. Bootstrap by sequential resampling. J. Statist. Plan. Inference 64.(1997).
Toan, N.V. On the asymptotic distribution of the bootstrap estimate with random resample size. Vietnam J. Math. 33:3, 261–270 (2005).
Toan, N.V. Rate of convergence in bootstrap approximations with random sample size. Acta Mathematica Vietnammica, 25. 161-179 (2000).
Toan, N.V. On Weak Convergence of the Bootstrap General Empirical Process with Random Resample Size. Vietnam J. Math, 42, 233–245 (2014).
D. A. Freedman. Bootstrap regression models. Ann. Statist. 9. (1981).
Toan, N.V. On bootstrapping regression and correlation models with random resample size. Vietnam J. Math, 37, 443–456 (2009).
Hữu, N.V and Dư, N.H. Phân tích thống kê và Dự báo. NXB ĐH Quốc Gia Hà Nội. (2003).
E. Mammen. When does bootstrap work. Springer-Verlag New York, Inc. (1992).
Toan, N.V. On weak convergence of the bootstrap empirical process with random resample size.Vietnam J. Math. 28:2, 153–158 (2000).
Downloads
Published
How to Cite
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Copyright © JTE.
