An interval feature discrete Grey-Markov model based on data distributions and applications

Data sequences represent complex changes in external environments or internal systems, and patterns of some index sequences have become more difficult to characterize using real number sequences. Considering the imbalances between regional development levels in terms of socioeconomic indicators, a discrete grey-Markov model based on interval distribution characteristics is proposed in this paper. Based on the classic discrete grey model, a difference coefficient sequence and median sequences of the upper and lower information domains, respectively, are introduced to establish differential equations for the representation and prediction of different development level trends. Furthermore, a combination of discrete grey models with the Markov model is used to modify the nonstationary data sequences and further improve the new model prediction accuracy. The parameter estimation and solution of the proposed model are organized using analytical methods, and the accuracy of the model is investigated. By fitting and predicting examples of private car ownership and milk production in representative unbalanced development regions, the results of the proposed model are compared with those of four other prediction models to verify its effectiveness. The results demonstrate that the prediction accuracy of the proposed model is better than those of other grey models in nonstationary data sequences. By capturing the interval features, this model achieves the goal of characterization of interval distribution information and further modification of the nonstationary series.(c) 2023 Elsevier Inc. All rights reserved.