On the bias of the score function of finite mixture models

We characterize the unbiasedness of the score function, viewed as an inference function for a class of finite mixture models. The models studied represent the situation where there is a stratification of the observations in a finite number of groups. We show that, under mild regularity conditions, the score function for estimating the parameters identifying each group's distribution is unbiased. We also show that if one introduces a mixture in the scenario described above, so that for some observations it is only known that they belong to some of the groups with a probability not in {0,1}, then the score function becomes biased. We argue then that under further mild regularity, the maximum likelihood estimate is not consistent. The results above are extended to regular models containing arbitrary nuisance parameters, including semiparametric models.

Automated summary

This study characterizes the unbiasedness of the score function in a class of finite mixture models, and shows that the score function becomes biased when certain unknown probabilities exist in the estimation of parameters for group distribution.