Bayesian estimation of random - intercept models using skew - laplace distribution

Bayesian estimation of random - intercept models using skew - laplace distribution

. Random-intercept models frequently used to analyze the correlated data. In fitting these models, the conventional assumption is that the error terms and the random intercepts are normally distributed. In many empirical applications, this assumption may be violated and thus the main concern of most recent studies is the use of alternative distributions. In this paper, we propose a new class of random-intercept models using the skewLaplace distribution. We then show by conducting simulation studies that the proposed model is flexible such that it can capture heavy tails, peakedness and skewness of the data generating process simultaneously. Since the statistical inference based on the marginal likelihood is complicated we present a Bayesian analysis by using the Markov chain Monte Carlo simulation. Finally, the application of proposed model is illustrated in the analysis of a real data set concerning the tax liability study.