您好,欢迎访问三七文档
1、多元线性回归(1)y,x1,x2,x3的相关系数矩阵相关性yx1x2x3Pearson相关性y1.000.556.731.724x1.5561.000.113.398x2.731.1131.000.547x3.724.398.5471.000Sig.(单侧)y..048.008.009x1.048..378.127x2.008.378..051x3.009.127.051.Ny10101010x110101010x210101010x310101010(2)系数a模型非标准化系数标准系数tSig.B的95.0%置信区间相关性B标准误差试用版下限上限零阶偏部分1(常量)-348.280176.459-1.974.096-780.06083.500x13.7541.933.3851.942.100-.9778.485.556.621.350x27.1012.880.5352.465.049.05314.149.731.709.444x312.44710.569.2771.178.284-13.41538.310.724.433.212三元线性回归方程:y=-348.280+3.754x1+7.101x2+12.447x3(3)模型汇总b模型RR方调整R方标准估计的误差Durbin-Watson1.898a.806.70823.441881.935因为R方=0.806,说明拟合的效果很好(4)Anovab模型平方和df均方FSig.1回归13655.37034551.7908.283.015a残差3297.1306549.522总计16952.5009在0.05的显著性水平下,y对自变量x1,x2,x3整体的线性回归效果是显著的。(5)系数a模型非标准化系数标准系数tSig.B的95.0%置信区间相关性B标准误差试用版下限上限零阶偏部分1(常量)-348.280176.459-1.974.096-780.06083.500x13.7541.933.3851.942.100-.9778.485.556.621.350x27.1012.880.5352.465.049.05314.149.731.709.444x312.44710.569.2771.178.284-13.41538.310.724.433.212在0.05的显著性水平下,y对x1、x3的线性关系不显著,对x2的线性关系也不太显著。(6)剔除x1、x3后再做回归分析Anovab模型平方和df均方FSig.1回归9049.33619049.3369.160.016a残差7903.1648987.895总计16952.5009系数a模型非标准化系数标准系数tSig.B的95.0%置信区间相关性B标准误差试用版下限上限零阶偏部分1(常量)-159.927129.711-1.233.253-459.042139.187x29.6893.201.7313.027.0162.30717.071.731.731.731此时只有x2这一个变量,F检验和t检验的结果表明:y对x2的线性关系显著(7)剔除前的回归系数95%的置信区间:1(-0.977,8.485)2(0.053,14.149)3(-13.415,38.310)剔除后的回归系数2的置信区间为(2.307,17.071)。(9)x01=75,x02=42,x03=3.1时0y=267.829,y0的95%置信区间为(204.4,331.2),近似区间为(20y),计算得到为(219.6,316.0)。(10)由剔除前的回归方程y=-348.280+3.754x1+7.101x2+12.447x3、剔除后的回归方程y=-159.927+9.689x2可知农业总产值X2对货运总量Y的影响程度比较大。
本文标题:1多元线性回归
链接地址:https://www.777doc.com/doc-3026380 .html