Journal of the Korean Data and Information Science Society

The Korean Data and Information Science Society (한국데이터정보과학회)
권호 표지

Bootstrap Confidence Intervals for Regression Coefficients under Censored Data

  • Cho, Kil-Ho (Department of Statistics, Kyungpook National University) ;
  • Jeong, Seong-Hwa (Department of Statistics, Kyungpook National University)
  • Published : 2002.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

Using the Buckley-James method, we construct bootstrap confidence intervals for the regression coefficients under the censored data. And we compare these confidence intervals in terms of the coverage probabilities and the expected confidence interval lengths through Monte Carlo simulation.

Keywords

References

  1. Annals of Statistics v.9 Some asymptotic theory for the bootstrap Bickel, P.;Freeman, D.
  2. Biometrica v.66 Linear regression with censored data Buckley, J.;James, I. R.
  3. Biometricka v.83 A note on Buckley-James, I. R. Currie, I. D.
  4. Annals of Statistics v.7 Bootstrap methods; Another look at the jackknife Efron, B.
  5. Canadian Journal of Statistics v.9 Nonparametric standard errors and confidence interval Efron, B.
  6. Journal of the American Statistical Association v.82 Better bootstrap confidence interval Efron, B.
  7. Annals of Statistics v.9 Bootstrapping regression model Freedman, D.
  8. Biometrika v.63 Least squares regression with censored data Miller, R. G.
  9. Annals of Statistics v.18 Estimation in a linear regression model with censored data Ritov, Y.