The asymptotic behavior of the MLE, Bayesian, minimum distance and some other estimators is studied in the asymptotics of large samples. It is shown that under regularity conditions these estimators are consistent, asymptotically normal and in a certain sense are asymptotically efficient. Then we describe what happens if some of the regularity conditions are not fulfilled (misspecißed and non identifiable models, null Fisher information etc.).
KeywordsLoss Function Regularity Condition Fisher Information Asymptotic Normality Bayesian Estimator
Unable to display preview. Download preview PDF.