摘要
以1999~2010年甘肃省CPI及影响因素的数据为基础,应用多元线性回归方法建立了甘肃省CPI的回归模型,并对其进行理论阐述和分析检验,得出该模型的校正可决系数R珚2=0.998 4,说明模型与数据的拟合度很高;通过运用方差分析和数据模拟检验方法对甘肃省CPI的回归模型进行了实际检验,结果显示该模型的精确度达到了99.462 6%,可用于甘肃省CPI的预测.
A multiple linear regression model of CPI was established for Gansu by collecting the relevant data from 1999 to 2010.It was tested from the theoretical and practical aspects.The adjusted determination coefficient of the model was obtained:2=0.998 4,showing a high degree of fitness between the model and data.The analysis of variance and data simulation test were applied to the inspection of Gansu actual CPI factors,indicating that the accuracy of the model was as high as 99.462 6%,so it could be used to forecast CPI in Gansu Province.
出处
《甘肃科学学报》
2012年第4期152-155,共4页
Journal of Gansu Sciences
基金
甘肃省自然科学基金(GS(2010)GXZ008)
关键词
CPI
多元回归分析
校正可决系数
方差分析
CPI
multiple regression analysis
adjusted determination coefficient
analysis of variance