摘要
针对随机波动性数据对灰色GM(1.1)模型预测精度的影响问题,提出了基于BX数据处理方法与马尔可夫链理论的灰色预测模型(BXGrey-Markov模型)。首先,引入BX数据生成法对原始数据进行处理,以弱化原始数据之间的随机性。在灰色预测方法的基础上,引入马尔可夫链预测理论,建立了灰色马尔可夫预测模型,它是将灰色预测模型与马尔可夫预测方法优化组合,灰色预测模型用于预测随机序列数据的总体发展趋势,而用马尔可夫链模型预测各数据在总体趋势下的随机波动性变化,得到随机时间数列趋势预测模型的解。通过上海市交通事故预测实际数据进行了验证表明:灰色马尔可夫预测模型预测精度高于GM(1,1)模型的预测精度。GM(1,1)模型的平均预测精度为42.29%,BX GM(1,1)-Markov模型的平均预测精度为86.9%。
Aimed at the effect of random fluctuation data on the precision of road traffic accidents, based on BX data generation method and Markov chain theory, the BXGrey-Markov model of road traffic accidents is established. First of all, it has introduced BX data generation method to process the original data sequence in order to weaken the randomness between the original dat~ Based on gray method of prediction, the Markov chains prediction method is presented and a gray-Maerkov model for prediction is proposed. The solution of the statistical model is got by the optimization combination of both gray prediction and Markov prediction, a gray system model is used to prediction the general tread of the data series changing status, and a time series Markov chain model is used to predict fluctuation of data change along the general trend. The example shows that the model can well and truly predict the evolvement and changing tread of the data series status. The precision of gray-Markov model for prediction is better than that of gray model, the predication accuracy which is based on the BXGM（ 1,1 ） Markov model is 86.9%, and the predication accuracy which is based on the GM（ 1,1 ） model is 42.29%.
出处
《机械设计与制造》
北大核心
2013年第12期127-130,共4页
Machinery Design & Manufacture
基金
上海市科委科技攻关计划项目(10230501500)
上海市重点学科建设项目资助(P1405)
