A note on an iterative algorithm for nonparametric estimation in biased sampling models

O. Davidov and G. Iliopoulos

A simple iterative estimation procedure for computing the non-parametric maximum likelihood estimator (NPMLE) in biased sampling models is discussed and studied in detail. A proof of convergence is provided. Numerical experiments show that the algorithm is significantly faster in terms of CPU time compared with the standard procedure.

Key words and phrases: biased sampling; nonparametric maximum likelihood; convex optimization.

Back to Home Page