Abstract

In the present paper, the smoothed estimators are proposed for estimating mixing proportion in a mixed model based on independent and not identically distributed (non-iid) random samples of the existing estimators proposed by Boes(1966) – James(1978) called BJ estimator here and which was constructed for estimating mixing proportion based on independent and identically distribution random samples. The proposed smoothed estimators are based on known “kernel function” as described in the introduction. The following results of the smoothed estimators are studied under the non-iid setup such as (a) its small sample behavior is compared with the unsmoothed version (BJ estimator) based on their mean square errors (MSEs) by using Monte-Carlo simulation and established the percentage gain in precision of smoothed estimator over its unsmoothed version measured interms of their mse. (b) its large sample properties such as almost surely (a.s.) convergence and asymptotic normality of these estimators are established in the present work.

Author: Rama Krishnaiah, Y. S., Manish Trivedi and Konda Satish

Received on: May, 2017

Accepted on: February, 2019