Nonparametric tests for homogeneity of the marginal distributions of paired data.

Bibliographic Details
Main Author:	Lam, Frederic Che-Yuen
Other Authors:	Chui, C. K. (degree committee member.), Parzen, E. (degree committee member.), Wehrly, T. E. (degree committee member.)
Format:	Thesis Book
Language:	English
Published:	1982.
Subjects:	Distribution (Probability theory) Monte Carlo method. Nonparametric statistics. Stochastic processes. Statistics Academic theses
Online Access:	Link to ProQuest Copy Link to OAKTrust copy ProQuest, Abstract

Description
Abstract:	In this dissertation, the Pyke-Shorack theorem for linear rank statistics is generalized to include testing for the equality of the marginal distributions in a bivariate population. This generalization is used to develop nonparametric tests which are not based on the differences of the pairs. By obtaining consistent estimators of the asymptotic variances, the tests are asymptotically distribution free. New test statistics are proposed for the location problem and for bivariate exponential data. Their powers and relative efficiencies are compared to classical tests under a variety of parent distributions.
Item Description:	"Major subject: Statistics." Typescript (photocopy). Vita.
Physical Description:	vii, 99 leaves ; 29 cm
Bibliography:	Includes bibliographical references (leaves 96-98).