Advances and Applications in Statistics
Volume 2, Issue 3, Pages 269 - 285
(December 2002)
|
|
PERFORMANCE OF THE PARAMETRIC BOOTSTRAP METHOD IN SMALL SAMPLE INTERVAL ESTIMATES
Denise Benton (U.S.A.) and K. Krishnamoorthy (U.S.A.)
|
Abstract: In this article, we
investigate the performance of the parametric
bootstrap (PB) method in interval estimation of
parameters in various statistical problems and
we present a numerical method to evaluate the
merit of PB inferential procedures. We consider
interval estimation of the Poisson mean, the
correlation coefficient of a bivariate normal
distribution, the largest eigenvalue of a normal
covariance matrix, the ratio of normal means,
and the common mean of several normal
populations. PB interval estimates are compared
with the ones based on exact methods and, in
situations where exact methods are unavailable,
their performance is evaluated by the numerical
method. This evaluation method is useful to
verify the validity of the PB method in complex
problems. |
Keywords and phrases: common mean, correlation coefficient, largest eigenvalue, Monte Carlo, Poisson distribution, ratio of two normal means. |
|
Number of Downloads: 7 | Number of Views: 1209 |
|