Asymptotically Distribution-Free Aligned Rank Order Tests for Composite Hypotheses for General Multivariate Linear Models,

Abstract

For general multivariate linear models, a composite hypothesis does not usually induce invariance of the joint distribution under appropriate groups of transformations, so that genuinely distribution-free tests do not usually exist. For this purpose, some aligned rank order statistics are incorporated in the proposal and study of a class of asymptotically distribution-free tests. Tests for the parallelism of several multiple regression surfaces are also considered. Finally the optimal properties of these tests are discussed. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1976
Accession Number
ADA036396

Entities

People

  • Madan L. Puri
  • Pranab Kumar Sen

Organizations

  • University of North Carolina at Chapel Hill

Tags

Communities of Interest

  • Energy and Power Technologies
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Air Force
  • Analysis Of Variance
  • Asymptotic Normality
  • Center Of Gravity
  • Composite Materials
  • Data Science
  • Distribution Theory
  • Estimators
  • Hypotheses
  • Information Science
  • Multivariate Analysis
  • Order Statistics
  • Rank Order Statistics
  • Statistical Algorithms
  • Statistics
  • Stochastic Processes
  • Theorems

Fields of Study

  • Mathematics

Readers

  • Neural Network Machine Learning.
  • Regression Analysis.