ON THE BOOTSTRAPPING HETEROSCEDASTIC REGRESSION MODELS
Abstract: The distributions of deviations of point estimators for parameters of interest
are essential in the evaluation of the efficiency of point estimators. The bootstrap
method suggested by B. Efron is one of the main methods directed at solving the
problem of producing distributions which mimic the unobserved distributions of
deviations.
The main object of this article is to study the asymptotic validity of the bootstrap in the
context of heteroscedastic regression models, using the central limit resampling theorem. In
the case of one-parameter linear regression, theoretical results are illustrated by an example
with simulated statistical data.
1991 AMS Mathematics Subject Classification: Primary: -; Secondary: -;
Key words and phrases: Bootstrap, heteroscedastic regression, resampling, ordinary
least squares estimates, central limit resampling theorem.