Award Date

1-1-2005

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Mathematical Sciences

First Committee Member

Hokwon Cho

Number of Pages

28

Abstract

In this thesis we deal with a sequential procedure for testing uniformity in a given multinomial distribution using inverse sampling. From a decision theoretic point of view, we devise an efficient stopping rule that satisfies a pre-determined P*-condition. Dirichlet distribution Type II will be primarily used for developing the inverse-type sequential procedure based on the decision theoretic point of view. We assume a non-zero cell probability (parameter) for given multinomial models. In particular, we will be focusing on the equal cell probability configuration (EPC) among all feasible cell configurations. One of the main goals is to find optimal sample sizes that resulted from a desirable probability level, the probability of correct decision P{CD}, in testing uniformity in multinomial models. As an illustration, "wheel of fortune" will be considered to fit the developed model. Finally, the developed procedure will be discussed via Monte Carlo experimentation.

Keywords

Models; Multinomial; Procedure; Sequential; Test; Uniformity

Controlled Subject

Statistics

File Format

pdf

File Size

747.52 KB

Degree Grantor

University of Nevada, Las Vegas

Language

English

Permissions

If you are the rightful copyright holder of this dissertation or thesis and wish to have the full text removed from Digital Scholarship@UNLV, please submit a request to digitalscholarship@unlv.edu and include clear identification of the work, preferably with URL.

Identifier

https://doi.org/10.25669/u90a-fcl8


Share

COinS