拟马尔可夫过程的磨光优化算法及应用

Smoothing Optimization Algorithm of Quasi-Markov Process and Its Application

通过对磨光法及马尔可夫过程的研究,马氏过程作为区间预测的一种方法,在很大程度上约束了它预测的科学性,另外,磨光法本身也是一种迭代的方法,对于拟合的精度还是难于控制,通过拟马尔可夫矩阵与磨光法相结合及优化工具,得到拟马尔可夫过程的磨光优化算法,实例表明:拟马尔可夫过程的磨光优化算法使修正磨光后的值逼近原数据值的程度较其它算法更好,而且,拟马尔可夫矩阵反应了从一种状态到另一种状态的转移程度,并且这种算法具有更好的推广和应用。

By studying polishing method and the Markov Process, Markov Process is a method for interval forecasting,that largely bound the scientific nature of its forecast, In addition, polishing method itself is an iterative method, the fitting accuracy is still difficult to control. In this paper, a smoothing optimization algorithm of quasi-Markov Process is got through combining quasi-Markov matrix and polishing method and optimization tools. The example expresses smoothing optimization algorithm of quasi-Markov Process is more efficient than the others in the degree of approximation of the original data values,moreover, quasi-Markov matrix reply transition degree from a state to another state and this algorithm has better promotion and application.