Asymptotic Efficiency of Maximum Likelihood Estimators Under Misspecified Models
Abstract
We illustrate with examples when and how maximum likelihood estimators continue to be asymptotically efficient even under misspecified models. Also, we provide a necessary and sufficient condition under which a subset of the vector of MLE's retains its asymptotic efficiency under misspecified models even though the MLE itself is not fully asymptotic efficient.
Keywords
Full Text:
PDFDOI: http://dx.doi.org/10.18187/pjsor.v8i3.525
Refbacks
- There are currently no refbacks.
Copyright (c)
Title
Asymptotic Efficiency of Maximum Likelihood Estimators Under Misspecified Models
Keywords
Asymptotic, Maximum Likelihood, Misspecified Models
Description
We illustrate with examples when and how maximum likelihood estimators continue to be asymptotically efficient even under misspecified models. Also, we provide a necessary and sufficient condition under which a subset of the vector of MLE's retains its asymptotic efficiency under misspecified models even though the MLE itself is not fully asymptotic efficient.
Date
2012-07-01
Identifier
Source
Print ISSN: 1816-2711 | Electronic ISSN: 2220-5810