Modified information criterion for detecting changes in skew slash distribution

Abstract Skew slash distribution is a distribution which considers both skewness and heavy tail. It is very useful in simulation studies and realistic in representing practical data due to its less peaks, especially in data sets that violate the assumption of normality. In this article, we propose a change-point detection procedure for skew slash distribution based on the modified information criterion (MIC). Meanwhile, we provide an estimation approach based on confidence distribution (CD) to measure the accuracy of change point location estimation. By comparing with the likelihood ratio test, the simulation results show that the MIC-based method performs better in terms of powers, the coverage probabilities and average lengths of confidence sets. In the end, we apply our proposed method to real data and locate the positions of the change points successfully.