Abstract
A procedure is developed for power analysis and sample size calculation for a class of complex testing problems regarding the largest binomial probability under a combination of treatments. It is shown that the asymptotic null distribution of the likelihood-ratio statistic is not parameter-free, but (Formula presented.) is a conservative asymptotic null distribution. A nonlinear Gauss-Seidel algorithm is proposed to uniquely determine the alternative for the power and sample size calculation given the baseline binomial probability. An example from an animal clinical trial is discussed.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 78-83 |
| Number of pages | 6 |
| Journal | Statistical Theory and Related Fields |
| Volume | 4 |
| Issue number | 1 |
| DOIs | |
| State | Published - Jan 2 2020 |
Keywords
- Asymptotic null distribution
- Gauss-Seidel
- binomial probability
- complex hypotheses
- logistic regression
- power
- sample size
- tests
ASJC Scopus subject areas
- Statistics and Probability
- Analysis
- Applied Mathematics
- Statistics, Probability and Uncertainty
- Computational Theory and Mathematics