Likelihood ratio tests and intersection-union tests
The likelihood ratio test (LRT) method is compared with the intersection-union test (IUT) method which is appropriate when the null hypothesis is expressed as a union of sets. Some relations between tests that result from these two methods are given. Conditions are obtained under which both methods yield the same test. Conditions also are given under which the size-\(\alpha\) IUT is uniformly more powerful that the size-\(\alpha\) LRT. The relationships are illustrated by investigating the problem of testing \(H_0:\min\{| \mu_1| ,| \mu_2| \}=0\) versus \(H_{a}:min\{| \mu_1| ,| \mu_2| \}>0\), where \(\mu_1\) and \(\mu_2\) are means of two normal populations. It is illustrated in an example how even more powerful tests may be found by careful consideration of the specific problem. For the entire collection see [Zbl 0990.62506].
- Two-stage likelihood ratio and union-intersection tests for one-sided alternatives multivariate mean with nuisance dispersion matrix
- Repeated challenge studies: A comparison of union-intersection testing with linear modeling
- Signed difference analysis: theory and application.
- The emperor's new tests. (With comments and a rejoinder).
- An appraisal of some aspects of statistical inference under inequality constraints
- More powerful tests for the sign testing problem
- Extending DFA-based multiple linear regression inference: application to acoustic impedance models
- On efficient exact experimental designs for ordered treatments
- Simultaneous test for the mean and variance with an application to the statistical process control
- \(P^{3}\) approach to intersection-union testing of hypotheses
- Two Bayesian/frequentist challenges for categorical data analyses
- Likelihood Inference for Unions of Interacting Discs
- Likelihood ratio tests for interval hypotheses with applications
- Bayesian inference about odds ratio structure in ordinal contingency tables
- Some correlations in intersection-union tests and their relationship with complete power
- Joint inference about the AUC and Youden index for paired biomarkers
- Global test for high-dimensional mediation: testing groups of potential mediators
- Optimal designs for testing pairwise differences: a graph-based game theoretic approach
- Nonparametric methods controlling the median of the false discovery proportion
This page was built for publication: Likelihood ratio tests and intersection-union tests
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3843173)
