Model-based clustering using a new multivariate skew distribution

Abstract Quite often real data exhibit non-normal features, such as asymmetry and heavy tails, present a latent group structure. In this paper, we first propose the multivariate skew shifted exponential normal distribution that can account for these characteristics. Then, use in finite mixture modeling framework. An EM algorithm is illustrated maximum-likelihood parameter estimation. We provide simulation study compares fitting performance of our model with those several alternative models. The comparison also conducted on dataset concerning log returns four cryptocurrencies.