Doctoral Dissertations

Abstract

"This dissertation investigates some selection and ranking procedures for two or more Weibull populations. In particular it studies a restricted subset selection procedure when the populations have a common shape parameter, a 'pick the one best ' procedure for different shapes, classification procedures for both cases, simultaneous and 'best population' confidence intervals for both cases, and, finally, the problem of selecting the one population with largest or smallest shape. Except for the last problem named, the populations are ranked by their reliability. All procedures employ estimates of the parameters of interest based on either uncensored or Type II censored samples. The estimates used are maximum likelihood and some simplified linear estimates. The probabilities of making a correct decision are found by expressing the event of a correct decision in terms of estimates from negative exponential parent populations and a 'preference zone' appropriate for the problem under consideration. Although it was inconvenient or impossible to exhibit a density for the estimates used, computer programs were written to determine the probability of a correct decision given the preference zone, or vice versa. The programs used Monte Carlo methods. For the restricted subset problem (using uncensored, linear estimates), it was found that the probability of a correct decision could be approximated closely by fitting the first two moments of the log of the estimate to a normal or log X2 distribution"-- Abstract, p. ii-iii

Advisor(s)

Patel, J. (Jagdish)

Committee Member(s)

Gillett, Billy E.
Wright, F. T.
Bain, Lee J., 1939-
Ho, C. Y. (Chung You), 1933-1998

Department(s)

Mathematics and Statistics

Degree Name

Ph. D. in Mathematics

Publisher

University of Missouri--Rolla

Publication Date

1978

Pagination

xv, 318 pages

Note about bibliography

Includes bibliographical references (pages 109-112)

Rights

© 1978 Joseph Victor Kingston, All rights reserved.

Document Type

Dissertation - Open Access

File Type

text

Language

English

Thesis Number

T 4421

Print OCLC #

6009383

Download

Included in

Mathematics Commons

Share

 
COinS