Multivariate Mixtures of Erlangs for Density Estimation Under Censoring and Truncation

Lifetime Data Analysis, 2016, 22(3), 429-455.

36 Pages Posted: 7 Jan 2015 Last revised: 17 May 2017

See all articles by Verbelen Roel

Verbelen Roel

KU Leuven

Katrien Antonio

KU Leuven; University of Amsterdam

Gerda Claeskens

KU Leuven - Department of Economics

Date Written: December 19, 2014

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 censored and truncated data. The effectiveness of the proposed algorithm, which has been implemented in R, is demonstrated on simulated as well as real data sets.

The Addendum for this paper are available at the following URL: http://ssrn.com/abstract=2593016

Keywords: Multivariate Erlang mixtures with a common scale parameter, Density estimation, Censored data, Truncated data, Expectation-maximization algorithm, Maximum likelihood.

Suggested Citation

Roel, Verbelen and Antonio, Katrien and Claeskens, Gerda, Multivariate Mixtures of Erlangs for Density Estimation Under Censoring and Truncation (December 19, 2014). Lifetime Data Analysis, 2016, 22(3), 429-455.. Available at SSRN: https://ssrn.com/abstract=2545391 or http://dx.doi.org/10.2139/ssrn.2545391

Verbelen Roel (Contact Author)

KU Leuven ( email )

Naamsestraat 69
B-3000 Leuven, Vlaams-Brabant 3000
Belgium

Katrien Antonio

KU Leuven ( email )

Leuven, Vlaams-Brabant

HOME PAGE: http://www.econ.kuleuven.be/katrien.antonio

University of Amsterdam ( email )

Roetersstraat 11
Amsterdam, 1018 WB
Netherlands

Gerda Claeskens

KU Leuven - Department of Economics ( email )

Leuven, B-3000
Belgium

Register to save articles to
your library

Register

Paper statistics

Downloads
41
Abstract Views
334
PlumX Metrics