Estimation Of Parameters In Finite Mixtures Of Distributions From Right Censored Data
Abstract
In the text the problem of estimation of parameters in finitemixtures of probability distributions in case of the presence of right censored data is treated. Only models with known number of components and observable component membership (complete mixture models) are studied. Possible estimation method of unknown parameters (parameters of components and mixing proportions)isdescribed. Specific theoretical and.numerical problems associated with this this type of the modelling are discussed with respect to the mixture models with complete data.A simulation studyis presented toillustrate properties of estimates and sensitivity of results on the proportion of censored data and the sample size. For the simulation10,000 samples with 500 and 1,000 observations from themixtures with two components of normal (symmetricdistribution) and lognormal (asymmetric distribution)distributions are generated. Results are given in the tablesand selected histograms of estimates of parameters areshown. All computations are made in the package R.
References
http://dx.doi.org/10.1002/0471721182
http://dx.doi.org/10.1093/biomet/88.2.535
http://dx.doi.org/ 10.1080/02664769922188
http://dx.doi.org/10.1016/j.jspi.2009.06.014
http://dx.doi.org/ 10.1111/j.1467 9469.2006.00501.x
Therneau, T. A Package for Survival Analysis in S. R package version 2.372. (2012).
A.F., Makov, U.E. Statistical analysis of
finite mixture distributions, Wiley Series in Probability and Statistics,John Wiley. 1985. www.researchjournals.co.uk
Copyright information
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (Creative Commons Attribution License 3.0 - CC BY 3.0) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).