A Predictive Analytics Approach Based on Grey-Markov for Digital Economy

Abstract

This paper presents a predictive analytics approach based on Grey-Markov for digital economy. It selects the cumulative value data of consumer goods retail sales in China from February to December 2020. Firstly, the GM(1, 1) model is established to forecast retail sales of consumer goods for each month of 2020. Markov model is used to divide three transition states according to the prediction error. Then the cumulative values of February and March 2021 are predicted by GM(1, 1), and the prediction results are corrected by Markov chain model. Through comparing the prediction results of GM(1, 1) model and GM(1, 1)-Markov model, it is found that the prediction accuracy of GM(1, 1)-Markov model is higher, which indicates that the GM(1, 1)-Markov model has certain practical significance for the short-term prediction of retail sales of consumer goods.