Accession Number : ADA262552


Title :   Modified Goodness-of-Fit Tests for the Weibull Distribution


Descriptive Note : Master's thesis


Corporate Author : AIR FORCE INST OF TECH WRIGHT-PATTERSONAFB OH SCHOOL OF ENGINEERING


Personal Author(s) : Yucel, Erol


Full Text : https://apps.dtic.mil/dtic/tr/fulltext/u2/a262552.pdf


Report Date : Mar 1993


Pagination or Media Count : 90


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....Weibull distribution, Goodness-of-fitness, Monte Carlo simulation, Minimum distance, Maximum likelihood.


Descriptors :   *STATISTICAL DISTRIBUTIONS , *GOODNESS OF FIT TESTS , SIMULATION , PARAMETERS , STATISTICS , SCALE , POWER


Subject Categories : Statistics and Probability


Distribution Statement : APPROVED FOR PUBLIC RELEASE