Sign in
Maximum Likelihood Estimation: for the Multinomial Distribution
Journal article   Peer reviewed

Maximum Likelihood Estimation: for the Multinomial Distribution

Charles S Davis and Michael P Jones
Teaching statistics, Vol.14(3), pp.9-11
09/1992
DOI: 10.1111/j.1467-9639.1992.tb00231.x

View Online

Abstract

In a first course in mathematical statistics or categorical data analysis, the maximum likelihood estimators of the parameters of the multinomial distribution are often “derived” using calculus. However, the important step of verifying that the resulting estimators indeed maximize the likelihood is omitted or done incorrectly. In this paper, two simple methods of obtaining the maximum likelihood estimates are presented. Copyright © 1992, Wiley Blackwell. All rights reserved
Teaching: Arithmetic-geometric mean inequality: Negative definite matrix

Details

Metrics

7 Record Views