论文信息 - Bootstrap approximation of nearest neighbor regression function estimates

Bootstrap approximation of nearest neighbor regression function estimates

Let (X, Y) be a random vector in the plane and denote by m(x) = (YX = x) the corresponding regression function. We show that the bootstrap approximation for the distribution of a smoothed nearest neighbor estimate of m(x) is valid. Also we compare, by Monte Carlo, confidence intervals which are obtained from both the normal and the bootstrap approximation.

Gerhard Dikta | G. Dikta

[1] Winfried Stute,et al. THE OSCILLATION BEHAVIOR OF EMPIRICAL PROCESSES , 1982 .

[2] K. Singh,et al. On the Asymptotic Accuracy of Efron's Bootstrap , 1981 .

[3] D. Freedman. Bootstrapping Regression Models , 1981 .

[4] W. Härdle,et al. Bootstrapping in Nonparametric Regression: Local Adaptive Smoothing and Confidence Bands , 1988 .

[5] J. Kiefer,et al. Asymptotic Minimax Character of the Sample Distribution Function and of the Classical Multinomial Estimator , 1956 .

[6] Richard L. Wheeden. Measure and integral , 1977 .

[7] W. Stute. Asymptotic Normality of Nearest Neighbor Regression Function Estimates , 1984 .

[8] B. Efron. The jackknife, the bootstrap, and other resampling plans , 1987 .

[9] Shie-Shien Yang,et al. Linear Functions of Concomitants of Order Statistics with Application to Nonparametric Estimation of a Regression Function , 1981 .

[10] Winfried Stute,et al. ON ALMOST SURE CONVERGENCE OF CONDITIONAL EMPIRICAL DISTRIBUTION FUNCTIONS , 1986 .