
Expectation-maximization estimators for incompletely observed data
Author(s) -
Vladimir Vasić
Publication year - 2004
Publication title -
economic annals/ekonomski anali
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.148
H-Index - 12
eISSN - 1820-7375
pISSN - 0013-3264
DOI - 10.2298/eka0461165v
Subject(s) - maximization , expectation–maximization algorithm , estimator , maximum likelihood , variety (cybernetics) , computation , mathematical optimization , mathematics , restricted maximum likelihood , maximum likelihood sequence estimation , computer science , statistics , algorithm
Expectation-maximization is a broadly applicable approach to the iterative computation of maximum likelihood estimates. Each iteration of expectation-maximization method consists of two steps: the expectation step and the maximization step. Expectation-maximization method is useful in a variety of problems where the maximum likelihood estimates are very difficult to find. The basic idea of expectation-maximization method is to relate incomplete data problems to complete data problems where estimation by maximum likelihood method is much simpler