相关分析及非线性回归上机实验.pptVIP

因为 data ex;input x y@@; cards; 2 0.3 4 0.86 6 1.73 8 2.2 10 2.47 12 2.67 14 2.8 ; proc nlin; model y=k/(1+a*exp(-b*x)); parms a=19.614 b=0.52 k=2.827; der.a=-k*exp(-b*x)/(1+a*exp(-b*x))**2; der.b=k*a*x*exp(-b*x)/(1+a*exp(-b*x))**2; der.k=1/(1+a*exp(-b*x)); run; Iter k a b Squares 0 2.8270 19.6140 0.5200 0.0618 1 2.7380 19.3199 0.5541 0.0322 2 2.7322 20.8544 0.5728 0.0307 3 2.7299 21.1471 0.5758 0.0307 4 2.7295 21.1740 0.5761 0.0307 5 2.7294 21.1789 0.5762 0.0307 6 2.7294 21.1797 0.5762 0.0307 7 2.7294 21.1799 0.5762 0.0307 故所求的Logistic曲线回归方程为 lyy=5.47789 Q=0.0307 完成习题 P74 第1,2题 P121 后面习题 * * 第二讲 相关分析 非线性回归 * * 例 甘薯实生苗栽培试验中薯块重 x1 ( g ) 、块根粗 x2 (mm)、单株结薯数 x3 及单株产量 y (g ) 的12组观测值如下表，求三种相关系数。 组号 x1 x2 x3 y 1 4.0 7.25 4.7 416.7 2 4.0 7.25 3.3 75.0 3 4.5 8.5 6.7 91.7 4 4.0 5.7 4.4 166.7 5 6.0 9.0 2.0 25.0 6 0.5 4.25 2.0 25.0 7 0.5 4.0 2.5 75.0 8 2.0 6.0 3.0 50.0 9 2.0 5.0 3.7 166.7 10 0.5 4.0 3.0 50.0 11 1.0 4.0 2.5 75.0 12 2.0 7.0 5.0 300.0 * * Data ex; input x1-x4@@; Cards; 4.0 7.25 4.7 416.7 4.0 7.25 3.3 75.0 4.5 8.5 6.7 91.7 4.0 5.7 4.4 166.7 6.0 9.0 2.0 25.0 0.5 4.25 2.0 25.0 0.5 4.0 2.5 75.0 2.0 6.0 3.0 50.0 2.0 5.0 3.7 166.7 0.5 4.0 3.0 50.0 1.0 4.0 2.5 75.0 2.0 7.0 5.0 300.0 ; proc corr;var x1-x4; proc corr;var x1 x2;partial x3 x4; proc corr;var x1 x3;partial x2 x4; proc corr;var x1 x4;partial x2 x3 ; proc corr;var x2 x3;partial x1 x4; proc corr;var x2 x4;partial x1 x3; proc corr;var x3 x4;partial x1 x2; proc reg;model x4=x1-x3; proc reg;model x3=x1 x2 x4; proc reg;model x2=x1 x3 x4; proc reg;model x1=x2-x4;run; (变量的记号x4也可以为y) * * 例中三种相关系数总表 j1 j2 x1 x2 x