On the asymptotic distribution of likelihood ratio test when parameters lie on the boundary

Abstract

The paper discusses statistical inference dealing with the asymptotic theory of likelihood ratio tests when some parameters may lie on boundary of the parameter space. We derive a closed form solution for the case when one parameter of interest and one nuisance parameter lie on the boundary. The asymptotic distribution is not always a mixture of several chi-square distributions. For the cases when one parameter of interest and two nuisance parameters or two parameters of interest and one nuisance parameter are on the boundary, we provide an explicit solution which can be easily computed by simulation. These results can be used in many applications, e.g. testing for random effects in genetics. Contrary to the claim of some authors in the applied literature that use of chi-square distribution with degrees of freedom as in case of interior parameters will be too conservative when some parameters are on the boundary, we show that when nuisance parameters are on the boundary, that approach may often be anti-conservative.Keywords Likelihood ratio test · Nuisance parameters on the boundary · One-sided tests · Parameters of interest on the boundary · Quadratic forms

Inference of Gene Flow in the Process of Speciation: An Efficient Maximum-Likelihood Method for the Isolation-with-Initial-Migration Model

“…A list of sufficient regularity conditions for this result can be found, for example, in Casella and Berger (2001, p. 516). One of them is clearly not met in the present case: in the pairwise comparison of some of our models, every point of ${\Phi }_0$ is a boundary point of $\Phi .$ In other words, if $H_0$ is true, the vector of true parameters $\phi *\in {\Phi }_0,$ whichever it might be, is on the boundary of $\Phi .$ This irregularity is present, for example, when $M_1=M_2=0$ according to $H_0$ and $M_1,$ $M_2\in \left[0,\mathrm{\infty }\right)$ according to $H_1.$ The problem of parameters on the boundary has been the subject of articles such as Self and Liang (1987) and Kopylev and Sinha (2011). The limiting distribution of the likelihood-ratio test statistic under this irregularity has been derived in these articles, but only for very specific cases.…”

Efficient Maximum-Likelihood Inference For The Isolation-With-Initial-Migration Model With Potentially Asymmetric Gene Flow

“…The second irregularity, that is, the problem of having parameters on the boundary, has been the subject of papers such as Self and Liang (1987) and Kopylev and Sinha (2011). The limiting distribution of the likelihood ratio test statistic under this irregularity has been derived in these papers, but only for very specific cases.…”

Some New Aspects of Statistical Inference for Multistage Dose-Response Models with Applications

“…Kopylev and Sinha (2011) discussed the case when the parameter of interest and a nuisance parameter lie on the boundary. Here we consider the case when two nuisance paramaters lie on the boundary and the parameter of interest is an interior point.…”