LEAST SQUARES ESTIMATION IN FINITE PROCESSES,
Abstract
A consistent estimate of the transitional probability matrix of a finite Markov process is given in the case when at each point in time only the proportions of the sample in each state are known. It is shown that this estimate is asymtotically more efficient, in a sense defined in the paper, than previously considered estimates for this matrix. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- May 05, 1958
- Accession Number
- AD0606900
Entities
People
- Albert Madansky
Organizations
- RAND Corporation