CONSTRAINED MAXIMUM LIKELIHOOD ESTIMATION OF N STOCHASTICALLY ORDERED DISTRIBUTIONS
Abstract
The study considers the problem of determining step function maximum likelihood estimates for N stochastically ordered distributions, subject to the constraint that the estimates themselves must also be stochastically ordered. This problem arises, for example, in the context of reliability growth. Brunk, et al., has achieved a closed form solution for the case N = 2, but was unable to extend the results to the case N > 2. We shall present a new analytical method based on the Kuhn-Tucker optimality conditions for the equivalent concave program. For N = 2, the method yields the closed form solution of Brunk, and for N = or > 3, the method yields an efficient computational algorithm. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1968
- Accession Number
- AD0673232
Entities
People
- Arthur M. Geoffrion
Organizations
- University of California, Los Angeles