On the Existence and Uniqueness of the Maximum Likelihood Estimate of a Vector-Valued Parameter in Fixed-Size Samples
Timo Makelainen, Klaus Schmidt, George P. H. Styan
The Annals of Statistics · 1981 · Volume 9 · Issue 4
Abstract
The maximum likelihood estimate is shown to exist and to be unique if a twice continuously differentiable likelihood function is constant on the boundary of the parameter space and if the Hessian matrix is negative definite whenever the gradient vector vanishes. The condition of constancy on the boundary cannot be completely removed, cf. Tarone and Gruenhage (1975). The theory is illustrated with several examples.
Keywords
62F10, 62H12, constancy on the boundary, Hessian matrix, likelihood equations, likelihood surface, Multivariate analysis, Unimodality
Enable JavaScript for citation tools and additional paper interactions.