Regression type tests for parametric hypotheses based on sums of squared L-statistics (Q1085907)

A procedure is developed for testing the hypothesis \(H_ 0: \theta =\theta_ 0\) against the alternative \(H_ A: \theta \neq \theta_ 0\) for a continuous, univariate distribution depending on a parameter vector \(\theta\). The statistic used for the test is a sum of squared L- statistics that is asymptotically equivalent in distribution, under both the null hypothesis and local alternatives, to the generalized likelihood ratio statistic for testing \(H_ 0\).