A Likelihood Ratio Test Regarding Two Nested but Oblique Order-Restricted Hypotheses

A note on testing two-dimensional normal mean

“…Analogous phenomena are reported by several authors, Berger [1], Berger et al [2], and Warrack et al [11]. Analogous phenomena are reported by several authors, Berger [1], Berger et al [2], and Warrack et al [11].…”

Towards a Unified Theory of Inequality Constrained Testing in Multivariate Analysis

“…This is not necessarily so if we have two nested hypotheses both of which involve inequality constraints. A discussion of an interesting example of that type where the least favorable configuration occurs at 'infinity' is given by Warrack & Robertson (1984). In the analysis of structural models the situation is complicated further by the fact that the asymptotic covariance matrix r usually depends on the population value of the parameter vector.…”

Tests for Monotonic Intensity in a Poisson Process

“…We say that C 0 and C a are non-oblique if this iterative projection property holds for every α. The terminology of non-obliqueness was first used by Warrack & Robertson (1984) and later studied by Menendez, Rueda & Salvador (1992), and Hu & Wright (1994). When C 0 and C a are non-oblique, the LRT statistic (3.1) is of D(x, ↔ x ) type, where ↔ x is the isotonic regression of x.…”