基于改善我国成品油定价机制的数学模型  

Mathematical Model Based on Improving China's Refined Oil Pricing Mechanism

作　　者：冉凯[1] 赵辛喆[2] 

机构地区：[1]西安文理学院数学与计算机工程学院,西安710065 [2]西北工业大学理学院,西安710129

出　　处：《西安文理学院学报（自然科学版）》2013年第1期21-26,共6页Journal of Xi’an University(Natural Science Edition)

基　　金：西安市科技计划资助项目(CXY1134WL08);西安市社会科学规划基金资助项目(文理专项)(12WL28)

摘　　要：针对目前我国成品油定价机制改善的问题建立相应的数学模型.分别采用分类评定模型(logit)进行条件概率估计、最小二乘法拟合等方法、得出成品油价格和国内居民消费价格指数(CPI)的相关性、倒挂问题和迟滞性,并提出改善措施.依据国内外成品油的价格及政策,利用多元回归模型分4个阶段分析了成品油价格.得出以10天为调价周期,2%为调价幅度红线的相关性最高(0.999 9的相关性).此外还根据目前成品油定价机制中的其他问题提出了合理的建议.In an effort to improve China＇ s refined oil pricing mechanism, the corresponding mathematical model is built. We adopt the classification assessment model （ logit ） for estima-ting the conditional probability. With the method of least squares, we illustrate the correlation, inconsistence and hysteresis between refined oil price and domestic consumer price index { CPI }. Based on domestic and international oil statistics and the related policies, we adopt piecewise multivariate regression model to analyze the four stages of oil price changes since 2007. We conclude that the price adjustment cycle is 10 days and pricing red line at 2% shows the highest correlation （ 0. 999 9 ）. In addition, we put forward some proposals to solve the problems of domestic refined oil pricing.

关 键 词：成品油定价机制 分段多元回归模型 迟滞性 相关性检验 

分 类 号：F206[经济管理—国民经济] F123.9