Prediction and analysis of domestic water consumption based on optimized grey and Markov model

Abstract With the rapid development of urbanization and continuous improvement living standards, China's domestic water consumption shows a growing trend. However, in some arid deficient areas, shortage resources is crucial factor affecting regional economic population growth. Therefore, it essential to reliably predict future data region. Aiming at problems poor prediction accuracy overfitting non-growth series traditional grey prediction, this paper uses residual model combined with Markov chain correction consumption. Based on theory established. Combined state transition matrix, value modified, applied Shaanxi Province from 2003 2019. The fitting results show that grade improved “good”. This dynamic unbiased can eliminate inherent error GM (1,1) model, improve accuracy, have better reliability, provide new method for prediction.