An exact multinomial test for equivalence

2009/03/01 by Jesse Frey · 1 voice
Mathematics · #Point processes and geometric inequalities #Statistical Methods and Bayesian Inference #Statistical Methods in Clinical Trials

paper · doi:10.1002/cjs.10000

openalex publication_date 2009/03/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/05/21

Abstract

Abstract Existing equivalence tests for multinomial data are valid asymptotically, but the α level is not properly controlled for small and moderate sample sizes. We resolve this difficulty by developing an exact multinomial test for equivalence and an associated confidence interval procedure. We also derive a conservative version of the test that is easy to implement even for very large sample sizes. Both tests use a notion of equivalence that is based on the cumulative distribution function, with two probability vectors being considered equivalent if their partial sums never differ by more than some specified constant. We illustrate the methods by applying them to Weldon's dice data, to data on the digits of π , and to data collected by Mendel. The Canadian Journal of Statistics 37: 47–59; © 2009 Statistical Society of Canada

Discussions

• I will start with what you are testing (does an unfair dice follow a hypothesized distribution), and how to test this. What you have is a multinomial distribution , and indeed a common test for this s [stackexchange, 3 points]

Related

• Testing equivalence of multinomial distributions -- a constrained bootstrap approach
• Two-Sample Test for Sparse High Dimensional Multinomial Distributions
• Pure Significance Tests for Multinomial and Binomial Distributions: the Uniform Alternative
• A Test for Equality of Distributions in High Dimensions
• A refinement of the multinomial distribution with application