The Annals of Statistics
- Ann. Statist.
- Volume 1, Number 5 (1973), 888-901.
On the Measurability and Consistency of Maximum Likelihood Estimates for Unimodal Densities
This paper is concerned with maximum likelihood estimates for a large class of families of unimodal densities. The existence of measurable maximum likelihood estimates and the consistency of asymptotic maximum likelihood estimates are proved. By counterexamples it is shown that the conditions which are sufficient for consistency cannot be removed without compensation.
Ann. Statist., Volume 1, Number 5 (1973), 888-901.
First available in Project Euclid: 12 April 2007
Permanent link to this document
Digital Object Identifier
Mathematical Reviews number (MathSciNet)
Zentralblatt MATH identifier
Primary: 28A20: Measurable and nonmeasurable functions, sequences of measurable functions, modes of convergence
Secondary: 62G05: Estimation 54A10: Several topologies on one set (change of topology, comparison of topologies, lattices of topologies) 54E45: Compact (locally compact) metric spaces
Unimodal density the mode of a unimodal density upper and lower semicontinuity Levy metric convergence in the mean pointwise convergence compact and locally compact metric space weak topoloty and topology induced by the supremum-metric on families of probability measures asymptotic maximum likelihood estimates existence of measurable estimates strong consistency
Reiss, Rolf-Dieter. On the Measurability and Consistency of Maximum Likelihood Estimates for Unimodal Densities. Ann. Statist. 1 (1973), no. 5, 888--901. doi:10.1214/aos/1176342509. https://projecteuclid.org/euclid.aos/1176342509