Power of the likelihood ratio test in covariance structure analysis
β Scribed by Albert Satorra; Willem E. Saris
- Book ID
- 112726728
- Publisher
- Springer
- Year
- 1985
- Tongue
- English
- Weight
- 673 KB
- Volume
- 50
- Category
- Article
- ISSN
- 0033-3123
No coin nor oath required. For personal study only.
π SIMILAR VOLUMES
Let \(\boldsymbol{W}\) be a \(p \times p\) matrix distributed according to the Wishart distribution \(W_{p}(n, \boldsymbol{\Phi})\) with \(\boldsymbol{\Phi}\) positive definite and \(n \geqslant p\). Let \(\left(m / \sigma^{2}\right) g\) be distributed according to the chi-squared distribution \(\ch
We prove that the power function of the likelihood ratio test for MANOVA attains its minimum when the rank of the location parameter matrix G decreases from s to 1. This provides a theoretical justification of a result that is known in the literature based only on numerical studies.