Semiparametric inference in mixture models with predictive recursion marginal likelihood
Preprint
Abstract
Predictive recursion is an accurate and computationally efficient algorithm for nonparametric estimation of mixing densities in mixture models. In semiparametric mixture models, however, the algorithm fails to account for any uncertainty in the additional unknown structural parameter. As an alternative to existing profile likelihood methods, we treat predictive recursion as a filter approximation to fitting a fully Bayes model, whereby an approximate marginal likelihood of the structural parameter emerges and can be used for inference. We call this the predictive recursion marginal likelihood. Convergence properties of predictive recursion under model mis-specification also lead to an attractive construction of this new procedure. We show pointwise convergence of a normalized version of this marginal likelihood function. Simulations compare the performance of this new marginal likelihood approach that of existing profile likelihood methods as well as Dirichlet process mixtures in density estimation. Mixed-effects models and an empirical Bayes multiple testing application in time series analysis are also considered.Keywords
All Related Versions
- Version 1, 2011-06-16, ArXiv
- Published version: Biometrika, 98 (3), 567.
This publication has 15 references indexed in Scilit:
- Bootstrap consistency for general semiparametric M-estimationThe Annals of Statistics, 2010
- Nonparametric Bayesian multiple testing for longitudinal performance stratificationThe Annals of Applied Statistics, 2009
- Consistency of a recursive estimate of mixing distributionsThe Annals of Statistics, 2009
- Asymptotic properties of predictive recursion: Robustness and rate of convergenceElectronic Journal of Statistics, 2009
- Stochastic Approximation and Newton’s Estimate of a Mixing DistributionStatistical Science, 2008
- Oracle and Adaptive Compound Decision Rules for False Discovery Rate ControlJournal of the American Statistical Association, 2007
- An Estimation Method for the Semiparametric Mixed Effects ModelBiometrics, 1999
- A recursive algorithm for nonparametric analysis with missing dataBiometrika, 1999
- Strong Laws for Martingale Differences and Independent Random VariablesJournal of Theoretical Probability, 1998
- A semiparametric Bayesian model for randomised block designsBiometrika, 1996