Premium
Calculating the Efficiency of Maximum Quasilikelihood Estimation
Author(s) -
Hill Joe R.,
Tsai ChihLing
Publication year - 1988
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.2307/2347341
Subject(s) - estimation , statistics , mathematics , econometrics , economics , management
SUMMARY Two instances for which maximum quasilikelihood estimation might be considered a viable alternative to more standard methods are considered. The first situation concerns certain data sets for which the statistician would like to assume nothing more about the distribution than an approximate mean‐variance relationship, [μ, V(μ)]. Transformation and quasilikelihood methods are compared for this case. The second situation involves models where the calculation of maximum likelihood estimates is complicated. In particular, mixture models which arise naturally when applying empirical Bayes or random effects methods are explored. Few general theoretical results are given; instead, the basic ideas are illustrated by clear examples.