Improved grey‐Markov chain algorithm for forecasting

Abstract

Purpose

The purpose of this paper is to present a method to accurately forecast the tendency of the gross amount of energy sources consumption of the country and construct a new kind of algorithm for forecasting that synthesizes the advantages of the grey model, Markov chains, and least square method.

Design/methodology/approach

With the application of this new algorithm, this paper have forecasted the trend of the gross amount of energy sources consumption of the country and come to the conclusions that the new algorithm is more precise than the grey model. It is proved that the improved grey‐Markov chain algorithm is effective and can be used by authorities to make decision.

Findings

It was found that combining the grey model, Markov chains, and least square method, can be a new algorithm for forecasting the trendency of the gross amount of energy sources consumption.

Research limitations/implications

The new algorithm is only suitable for the short‐term forecast.

Originality/value

The grey forecasting method reflects the overall tendency of primitive data sequence of the gross amount of energy source, and the Markov chain forecasting method reflects the effect of the random fluctuation. The least square method reflects the tendency of increase. The new algorithm is more precise than the grey model.

Citation

Zhijun, L., Weiwei, W. and Mian‐yun, C. (2009), "Improved grey‐Markov chain algorithm for forecasting", Kybernetes, Vol. 38 No. 3/4, pp. 329-338. https://doi.org/10.1108/03684920910944010