EM algorithm for mixture of skew-normal distributions fitted to grouped data.

Mahdi Teimouri
Author Information
  1. Mahdi Teimouri: Faculty of Science and Engineering, Department of Statistics, Gonbad Kavous University, Gonbad Kavous, Iran. ORCID

Abstract

Grouped data are frequently used in several fields of study. In this work, we use the expectation-maximization (EM) algorithm for fitting the skew-normal (SN) mixture model to the grouped data. Implementing the EM algorithm requires computing the one-dimensional integrals for each group or class. Our simulation study and real data analyses reveal that the EM algorithm not only always converges but also can be implemented in just a few seconds even when the number of components is large, contrary to the Bayesian paradigm that is computationally expensive. The accuracy of the EM algorithm and superiority of the SN mixture model over the traditional normal mixture model in modelling grouped data are demonstrated through the simulation and three real data illustrations. For implementing the EM algorithm, we use the package called ForestFit developed for R environment available at https://cran.r-project.org/web/packages/ForestFit/index.html.

Keywords

References

  1. J Behav Med. 2002 Jun;25(3):293-315 [PMID: 12055779]
  2. Biometrics. 1988 Jun;44(2):571-8 [PMID: 3390510]
  3. Biometrics. 2000 Dec;56(4):1249-55 [PMID: 11129487]
  4. Psychol Methods. 2003 Sep;8(3):338-63 [PMID: 14596495]
  5. Psychon Bull Rev. 2008 Aug;15(4):692-712 [PMID: 18792497]
  6. Stat Med. 2009 Nov 30;28(27):3363-85 [PMID: 19731223]
  7. Funct Integr Genomics. 2008 Aug;8(3):181-6 [PMID: 18210172]
  8. Math Biosci. 2014 Mar;249:60-74 [PMID: 24491286]
  9. Popul Stud (Camb). 2016 Jul;70(2):259-72 [PMID: 27031180]

Word Cloud

Created with Highcharts 10.0.0dataEMalgorithmmixturegroupedskew-normalmodelstudyuseSNsimulationrealdistributionsGroupedfrequentlyusedseveralfieldsworkexpectation-maximizationfittingImplementingrequirescomputingone-dimensionalintegralsgroupclassanalysesrevealalwaysconvergesalsocanimplementedjustsecondsevennumbercomponentslargecontraryBayesianparadigmcomputationallyexpensiveaccuracysuperioritytraditionalnormalmodellingdemonstratedthreeillustrationsimplementingpackagecalledForestFitdevelopedRenvironmentavailablehttps://cranr-projectorg/web/packages/ForestFit/indexhtmlfittedmaximumlikelihoodmethoddistribution

Similar Articles

Cited By

No available data.