Each week, we will be publishing layman’s abstracts of new articles from our prestigious portfolio of journals in statistics. The aim is to highlight the latest research to a broader audience in an accessible format.
The article featured today is from the Canadian Journal of Statistics, with the full article now available to read in early view here.
Chen, J., Li, P. and Liu, G. (2020), Homogeneity testing under finite location‐scale mixtures. Can J Statistics. doi:10.1002/cjs.11557
If a population contains several related, but different subpopulations, the distribution of its character might show signs of mixture. On the other way, if a distribution shows signs of mixture, the population is likely made of several subpopulations. Confirming the presence of subpopulations is an important research problem in many scientific investigations. There is, therefore, a need for statistical methods to effectively quantify the evidence of the mixture based on random samples. This paper provides an EM-test solution to test for homogeneity for finite mixtures of location-scale family distributions. We work out the nonstandard limiting distributions accompanied by a numerical method for real data analysis. We confirm the proposed EM-test is effective by both simulation experiments and application examples.