应用回归析 第五章习题.doc

应用回归分析 第五章习题 5.10 (1)建立y对的线性回归方程 Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 5922.827 2504.315 2.365 .040 x2 4.864 2.507 .677 1.940 .081 x3 2.374 .842 .782 2.818 .018 x4 -817.901 187.279 -1.156 -4.367 .001 x5 14.539 147.078 .050 .099 .923 x6 -846.867 291.634 -.899 -2.904 .016 a. Dependent Variable: y 由上可知，线性回归方程是： （2）用后退法选择变量 Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) -2530523.651 1053982.823 -2.401 .040 x2 -27.458 13.588 -3.823 -2.021 .074 x3 3.321 .797 1.094 4.169 .002 x4 -1506.217 324.836 -2.128 -4.637 .001 x5 212.489 146.255 .737 1.453 .180 x6 -477.930 284.609 -.507 -1.679 .127 x1 1304.787 542.184 5.104 2.407 .039 2 …………………………………………………………………… 3 (Constant) -445380.948 110447.795 -4.033 .002 x3 2.310 .457 .761 5.055 .000 x4 -971.882 174.101 -1.373 -5.582 .000 x6 -827.999 220.276 -.879 -3.759 .003 x1 232.202 56.138 .908 4.136 .002 a. Dependent Variable: y ANOVAd Model Sum of Squares df Mean Square F Sig. 1 Regression154 6 3306439.359 12.485 .001a Residual 2383520.783 9 264835.643 Total937 15 2 Regression033 5 3855922.807 13.104 .000b Residual 2942542.904 10 294254.290 Total938 15 3 Regression892 4 4687594.223 14.852 .000c Residual 3471780.046 11 315616.368 Total937 15 a. Predictors: (Constant), x1, x3, x6, x4, x5, x2 b. Predictors: (Constant), x1, x3, x6, x4, x2 c. Predictors: (Constant), x1, x3, x6, x4 d. Dependent Variable: y Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .945a .893 .821 514.622 2 .931b .868 .801 542.452 3 .919c .844 .787 561.797 a. Predictors: (Constant), x1, x3, x6, x4, x5, x2 b. Predictors: (Constant), x1, x3, x6, x4, x2 c. Predictors: (Constant), x1, x3, x6, x4 由上三表可知，用后退法选出的变量及其回归方程为： （3）用逐步回归法选择自变量 Mode