Estimating density function via constrained maximum likelihood method
- Univ. of California, Davis, CA (United States)
In this paper we discussed the nonparametric density estimation problem. We are most interested in an estimation which is {open_quotes}optimal{close_quotes} in some sense and can incorporate the prior information one knows about the density function. By applying the maximum likelihood method to the nonparametric density estimation problem and incorporating some type of prior information, say support of the density function, shape of the density function, etc, as constraints, we established an infinite dimensional constrained optimization model. We use the newly developed epi-analysis theory to the problem, we proved the consistency of constrained maximum likelihood estimators and in the case that the constraint set have the {open_quotes}uniformly approximation{close_quotes} property we proved the epi-distance convergence of the empirical problem to the limiting problem and hence establish the upper bound of some constrained maximum likelihood estimator by the epi-distance between the empirical problem and the limiting problem. Furthermore, numerical simulation examples have been implemented.
- OSTI ID:
- 35958
- Report Number(s):
- CONF-9408161-; TRN: 94:009753-0224
- Resource Relation:
- Conference: 15. international symposium on mathematical programming, Ann Arbor, MI (United States), 15-19 Aug 1994; Other Information: PBD: 1994; Related Information: Is Part Of Mathematical programming: State of the art 1994; Birge, J.R.; Murty, K.G. [eds.]; PB: 312 p.
- Country of Publication:
- United States
- Language:
- English
Similar Records
A maximum entropy method for MEG source imaging
Maximum-likelihood identification of structurally constrained covariances for initial-condition statistics