请教各位,本人是SPSS初学者,检验了变量存在多重共线性,就学习了用岭回归。
R-SQUARE AND BETA COEFFICIENTS FOR ESTIMATED VALUES OF K
K RSQ X1 X2 X3 X4 X5
______ ______ ________ ________ ________ ________ ________
.00000 .94630 .302600 .636939 -.101855 .571581 .145848
.05000 .94189 .040366 .636122 -.044788 .269722 .182954
.10000 .93668 .003241 .589625 -.018886 .228342 .206794
.15000 .93093 -.015640 .550983 -.000439 .211093 .219157
.20000 .92494 -.028426 .519082 .013566 .201592 .225565
.25000 .91886 -.038068 .492218 .024509 .195506 .228648
.30000 .91278 -.045719 .469164 .033212 .191174 .229757
.35000 .90673 -.051960 .449067 .040217 .187829 .229638
.40000 .90073 -.057140 .431317 .045907 .185077 .228730
.45000 .89480 -.061490 .415470 .050558 .182697 .227305
.50000 .88893 -.065171 .401194 .054377 .180560 .225536
.55000 .88312 -.068304 .388230 .057521 .178588 .223537
.60000 .87738 -.070980 .376381 .060113 .176732 .221388
.65000 .87169 -.073273 .365485 .062249 .174959 .219142
.70000 .86606 -.075238 .355416 .064005 .173248 .216837
.75000 .86048 -.076924 .346067 .065444 .171585 .214501
.80000 .85496 -.078368 .337353 .066615 .169962 .212153
.85000 .84948 -.079603 .329200 .067559 .168371 .209808
.90000 .84405 -.080655 .321547 .068310 .166808 .207476
.95000 .83867 -.081547 .314344 .068897 .165270 .205165
1.0000 .83334 -.082299 .307544 .069343 .163755 .202880
然后我选择了k=0.1
结果如下:
Run MATRIX procedure:
****** Ridge Regression with k = 0.10 ******
10 ** 4 X
Mult R .000096782
RSquare .000093668
Adj RSqu .000089711
SE 8.086802865
ANOVA table
df SS MS
Regress 5.000 7.74E+011 1.55E+011
Residual 8.000 5.23E+010 6.54E+009
F value Sig F
23.66961327 .00013416
--------------Variables in the Equation----------------
10 ** 6 X
B SE(B) Beta B/SE(B)
X1 .005215788 .137357226 .000000003 .000000038
X2 .000000801 .000000141 .000000590 .000005687
X3 -.277801082 1.321847648 -.000000019 -.000000210
X4 .000000014 .000000005 .000000228 .000002581
X5 .000000152 .000000088 .000000207 .000001721
Constant -.074157841 .143771775 .000000000 -.000000516
------ END MATRIX -----
R的平方太小太小了
请问是怎么回事 谢谢各位