Date of Award
3-1993
Document Type
Thesis
Degree Name
Master of Science
Department
Department of Operational Sciences
First Advisor
Albert H. Moore, PhD
Abstract
This research will produce a new modified Anderson-Darling and W* goodness of fit statistics for the three parameter Weibull Distribution when all parameters unknown and estimated by Maximum Likelihood-Minimum Distance combination. The critical values for each statistic, sample size and a levels 0. 20, 0.15, 0.10, 0.05 and 0.01 are produced. The Monte Carlo Simulation used 5000 repetitions for sample sizes of 10, 15, 20, 25, 30, 40, and 50 with the Weibull shape equals to 3, scale equals to 4 and location equals to 10. The power study is made for the same sizes as above with the hypothesized Weibull Distribution against 8 other distributions.
AFIT Designator
AFIT-GOR-ENS-93M-25
DTIC Accession Number
ADA262552
Recommended Citation
Yucel, Erol, "Modified Goodness-of-Fit Tests for the Weibull Distribution" (1993). Theses and Dissertations. 7220.
https://scholar.afit.edu/etd/7220
Comments
The author's Vita page is omitted.