stata是用RXridge iit fdi idc rc
结果如下:
RXridge: Estimated Sigma = .54503108
RXridge: Uncorrelated Components... Number of obs = 0
------------------------------------------------------------------------------
iit | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
c1 | .3399242 .1231436 2.76 0.028 .0487358 .6311127
c2 | .4622791 .1785618 2.59 0.036 .0400475 .8845107
c3 | 1.830634 .5211019 3.51 0.010 .5984235 3.062844
------------------------------------------------------------------------------
SPSS是这么做的
include '。。。\Ridge regression.sps'.
ridgereg enter = fdi idc rc
/dep = iit
/inc = 0.01.
出来的结果里,按照stata的意思,k值 Estimated Sigma = .54503108,我去找了一下k=0.54的spss结果
R-SQUARE AND BETA COEFFICIENTS FOR ESTIMATED VALUES OF K
K RSQ FDI IDC RC
______ ______ ________ ________ ________
.54000 .50119 .487471 .039942 .111934
.55000 .49869 .483360 .037293 .110209
这两个差的也太多了啊?
求大神解救。
是不是STATA里面的系数C1\C2\C3不是自变量的系数哇?