The Annals of Statistics

Bootstrapping the Maximum Likelihood Estimator in High-Dimensional Log-Linear Models

Wilhelm Sauermann

Full-text: Open access

Abstract

The notion of a bootstrap estimator of the distribution of the maximum likelihood estimator in log-linear models is defined for common sampling models. It is shown that the bootstrap estimator is consistent under assumptions which allow the dimension of the model to increase to infinity. Such an approach allows treatment of large, sparse contingency tables.

Article information

Source
Ann. Statist., Volume 17, Number 3 (1989), 1198-1216.

Dates
First available in Project Euclid: 12 April 2007

Permanent link to this document
https://projecteuclid.org/euclid.aos/1176347264

Digital Object Identifier
doi:10.1214/aos/1176347264

Mathematical Reviews number (MathSciNet)
MR1015146

Zentralblatt MATH identifier
0683.62025

JSTOR
links.jstor.org

Subjects
Primary: 62G05: Estimation
Secondary: 62H17: Contingency tables

Keywords
Bootstrap decomposable log-linear models sampling models model asymptotics sparse contingency tables

Citation

Sauermann, Wilhelm. Bootstrapping the Maximum Likelihood Estimator in High-Dimensional Log-Linear Models. Ann. Statist. 17 (1989), no. 3, 1198--1216. doi:10.1214/aos/1176347264. https://projecteuclid.org/euclid.aos/1176347264


Export citation