Developing a crisis model based on higher-order moments

Abstract

In this paper, the goal is to create a model for detecting a crisis situation using the example of financial time series. An analysis of fixed and cumulative central and raw moments has been carried out. When conducting the analysis by higher-order moments, it has been shown that the moments record the presence of deviations and extreme values in the distribution of the series. Based on the higher-order moments, a crisis indicator has been proposed.

Keywords

• Time series
• Statistical moment
• Point estimation
• Higher-order moments