In some biological experiments, it is quite common that laboratory subjects may be different in their patterns of susceptibility to a treatment. Meantime some commercial product must have been manufactured. In these situations,finite mixture model analysis becomes a useful tool. This article mainly applys MCMC technics to solve a mixture of distributions model on real data. The parameters of each component are estimated using the Gibbs sampler.And the results will be compared with the EM algorithm in Refer-ence[l]. At last determination of the number of components is also considered,via model selection using the Bayes factors.
|