Fitting Nonparametric and Semeparametric Models Seminar

Nonparametric and semiparametric models are valuable tools for solving many nasty problems. While the maximum likelihood approach is often favoured, in most cases their fitting has long been known to be computationally difficult, due to the estimation of a distribution function defined on an infinite-dimensional space. In this talk, I will describe some fast algorithms for fitting nonparametric and semeparametric mixtures and shape-restricted models, and discuss some of their practical applications.