运用变系数回归模型,根据1980-2014年数据,分析吉林省居民消费水平,并与普通线性回归结果进行比较。根据残差平方和与居民消费水平预测值,相比于线性模型,变系数回归模型在拟合精度和预测精度上都有所改善。
This paper used the variable coefficient model which is based on 35 years data from 1980 to 2014 year to fit the Jilin province residents'consumption level, and then compared the given data with the fitting of the linear regression model. The results demonstrated that the variable coefficient model has a better fitting effect.