A Smooth Nonparametric Estimate of a Mixing Distribution Using Mixtures of Gaussians

Abstract We propose a method of estimating mixing distributions using maximum likelihood over the class of arbitrary mixtures of Gaussians subject to the constraint that the component variances be greater than or equal to some minimum value h. This approach can lead to estimates of many shapes, with smoothness controlled by parameter h. We show that the resulting estimate will always be a finite mixture of Gaussians, each having variance h. The nonparametric maximum likelihood estimate can be viewed as a special case, with h = 0. The method can be extended to estimate multivariate mixing distributions. Examples and the results of a simulation study are presented.

[1]  J. Kiefer,et al.  CONSISTENCY OF THE MAXIMUM LIKELIHOOD ESTIMATOR IN THE PRESENCE OF INFINITELY MANY INCIDENTAL PARAMETERS , 1956 .

[2]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[3]  N. Laird.  Nonparametric Maximum Likelihood Estimation of a Mixing Distribution , 1978 .

[4]  C. Morris.  Parametric Empirical Bayes Inference: Theory and Applications , 1983 .

[5]  B. Lindsay.  The Geometry of Mixture Likelihoods: A General Theory , 1983 .

[6]  J. Phair,et al.  The Multicenter AIDS Cohort Study: rationale, organization, and selected characteristics of the participants. , 1987, American journal of epidemiology.

[7]  R. Carroll,et al.  Deconvolving kernel density estimators , 1987 .

[8]  A. Gallant,et al.  Semi-nonparametric Maximum Likelihood Estimation , 1987 .

[9]  P. Hall,et al.  Optimal Rates of Convergence for Deconvolving a Density , 1988 .

[10]  D. Bates,et al.  Newton-Raphson and EM Algorithms for Linear Mixed-Effects Models for Repeated-Measures Data , 1988 .

[11]  Diane Lambert,et al.  Generalizing Logistic Regression by Nonparametric Mixing , 1989 .

[12]  Cun-Hui Zhang.  Fourier Methods for Estimating Mixing Densities and Distributions , 1990 .

[13]  D. Bates,et al.  Nonlinear mixed effects models for repeated measures data. , 1990, Biometrics.

[14]  T. Louis,et al.  Smoothing the non-parametric estimate of a prior distribution by roughening: A computational study , 1991 .

[15]  J. Kalbfleisch,et al.  An Algorithm for Computing the Nonparametric MLE of a Mixing Distribution , 1992 .

[16]  Bruce G. Lindsay,et al.  Computer-assisted analysis of mixtures (C.A.MAN) statistical algorithms , 1992 .

[17]  T A Louis,et al.  Random effects models with non-parametric priors. , 1992, Statistics in medicine.

[18]  R. Douglas Martin,et al.  S-PLUS Version 3 , 1992 .

[19]  N. Breslow,et al.  Approximate inference in generalized linear mixed models , 1993 .

[20]  J. Chmiel,et al.  Tracking of markers and onset of disease among HIV-1 seroconverters. , 1993, Statistics in medicine.

[21]  P. Diggle,et al.  Semiparametric models for longitudinal data with application to CD4 cell numbers in HIV seroconverters. , 1994, Biometrics.