MOMENT ESTIMATORS FOR THE PARAMETERS OF A MIXTURE OF TWO BINOMIAL DISTRIBUTIONS
Abstract
The distribution of a chance variable Y is said to be a mixture of two binomial distributions if P Y=y =(ny) py1(1-p1)n-y+(1- )py2(1-p2)n-y for y=0,1,..., or n=0, otherwise where 0< <1, 0<p1<p2<1 and n 3 is integral. Estimators for , p1, and p2 are constructed on the basis of sample factorial moments computed on m observations from this distribution. The asymptotic joint distribution of the estimators and the asymptotic efficiency of the moment estimators relative to the maximum likelihood estimators are computed. The asymptotic efficiency is discussed as a function of the binomial parameter n. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 30, 1961
- Accession Number
- AD0259916
Entities
People
- Wallace R. Blischke