Rank-order tests for the parallelism of several regression surfaces (Q1063976)

For testing the hypothesis that several (s\(\geq 2)\) linear regression surfaces \(X_{ki}=\alpha_ k\beta_ kc_{ki}+Z_{ki}\) \((k=1,...,s)\) are parallel to one another, i.e., \(\beta_ 1=...=\beta_ s\), a class of rank-order tests are considered. The tests are shown to be asymptotically distribution-free, and their asymptotic efficiency relative to the general likelihood ratio test is derived. Asymptotic optimality in the sense of Wald is also discussed.