Diagnostic Parameter Determination for a Class of Three-Parameter Probability Distributions
Abstract
The problem of parameter estimation of the non-shifted hyper-Gamma class is being considered. The approach is based on an applicability criterion which provides the opportunity to determine the parameters by means of three equations derived from the first and second moments and an analytical approximation of the logarithm of the cumulative distribution function. The parameter estimation process requires the iterative solution of two equations. Examples are given to verify the efficiency of the proposed parameter determination method. Recently obtained results on maximum-likelihood parameter estimation for the hyper-Gamma class may turn out to be practically more reliable than those presented in this report, especially for non-smooth data.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1987
- Accession Number
- ADA203706
Entities
People
- H. P. Dudel
- S. H. Lehnigk
Organizations
- United States Army Aviation and Missile Command