PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
January 1, 1943Transactions of the American Mathematical Society2,298 citationsOpen Access

Tests of statistical hypotheses concerning several parameters when the number of observations is large

View Full Paper
AWAbraham WaldCity University of New York

Key Points

  • This work aims to evaluate tests of statistical hypotheses across multiple parameters with large sample sizes. It focuses on determining the optimum properties of various tests.
  • Reduction to multivariate normal distribution for analysis.
  • Evaluation of average power and constant power tests on families of surfaces.
  • Discussion on the likelihood ratio test's large sample distribution.
  • Identifies tests with uniformly best average power across hypothesis families.
  • Determines tests with best constant power for specific surfaces.
  • Analyzes the properties of the likelihood ratio test for large samples.

Abstract

Table of contents 1. Introduction.426 2. Assumptions on the density function/(x, 6).428 3. The joint limit distribution of 0".429 4. Reduction of the general problem to the case of a multivariate normal distribution. .433 5. Tests of simple hypotheses which have uniformly best average power over a family of surfaces.445 6. Tests of simple hypotheses which have best constant power on a family of surfaces. . .450 7. Most stringent tests of simple hypotheses.451 8. Definitions of "best" tests of composite hypotheses.453 9. Tests of linear composite hypotheses which have uniformly best average power over a family of surfaces.455 10.Tests of linear composite hypotheses which have best constant power on a family of surfaces.461 11.Most stringent tests of linear composite hypotheses.461 12.The general composite hypothesis.463 13.Optimum properties of the likelihood ratio test.470 14.Large sample distribution of the likelihood ratio.478 15.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Abraham Wald (1943) studied this question.

synapsesocial.com/papers/697a49aefc2ec97650bea216https://doi.org/10.1090/s0002-9947-1943-0012401-3
Ask AI
Helpful
Bookmark
Share
View Full Paper