Hi, Below are the average loglikelihood of mixture model over 5 components (k=5). The loglikelihood are learnt using EM over three kinds of distribution: Lognormal, Gamma, and Normal.
Since the average loglikelihood for Lognormal is the highest over Gamma and Normal. Is it safe to infer that Lognormal is the best distribution to fit the data? . LOGNORMAL k,loglik 1,-212.8506847 2,-205.5816305 3,-204.5489356 4,-203.0542678 5,-201.7842441 GAMMA 1,-968.587694 2,-955.1132214 3,-963.1998383 4,-967.1973333 5,-968.5640199 NORMAL 1,-1063.016157 2,-979.1302761 3,-952.2856567 4,-913.6550025 5,-881.6734428 Regards, Gundala Viswanath Jakarta-Indonesia ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
