Author:

Keywords:

Multivariate mixtures of Erlangs with a common scale parameter, Density estimation, Censored data, Expectation-maximization algorithm, Maximum likelihood, Science & Technology, Physical Sciences, Mathematics, Interdisciplinary Applications, Statistics & Probability, Mathematics, PHASE-TYPE DISTRIBUTIONS, QUARTER INFECTION TIMES, MAXIMUM-LIKELIHOOD, TRUNCATED DATA, LOSSES, MODELS, Expectation–maximization algorithm, Algorithms, Humans, Likelihood Functions, Multivariate Analysis, 0104 Statistics, 4905 Statistics

Abstract:

Multivariate mixtures of Erlang distributions form a versatile, yet analytically tractable, class of distributions making them suitable for multivariate density estimation. We present a flexible and effective fitting procedure for multivariate mixtures of Erlangs, which iteratively uses the EM algorithm, by introducing a computationally efficient initialization and adjustment strategy for the shape parameter vectors. We furthermore extend the EM algorithm for multivariate mixtures of Erlangs to be able to deal with randomly censored and fixed truncated data. The effectiveness of the proposed algorithm is demonstrated on simulated as well as real data sets.