1、分别用前进法和后推法 回归结果如下
. stepwise reg y x1-x5,pe(0.05)
begin with empty model
p = 0.0000 < 0.0500 adding x1
p = 0.0000 < 0.0500 adding x2
Source | SS df MS Number of obs = 14
-------------+------------------------------ F( 2, 11) = 997.41
Model | 48033207.2 2 24016603.6 Prob > F = 0.0000
Residual | 264869.664 11 24079.0604 R-squared = 0.9945
-------------+------------------------------ Adj R-squared = 0.9935
Total | 48298076.8 13 3715236.68 Root MSE = 155.17
------------------------------------------------------------------------------
y | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
x1 | .0409783 .0026847 15.26 0.000 .0350693 .0468874
x2 | 5.142659 .670864 7.67 0.000 3.666098 6.619221
_cons | -3489.548 265.1268 -13.16 0.000 -4073.088 -2906.008
------------------------------------------------------------------------------
. stepwise reg y x1-x5,pr(0.05)
begin with full model
p = 0.3436 >= 0.0500 removing x4
p = 0.1334 >= 0.0500 removing x5
p = 0.0766 >= 0.0500 removing x3
Source | SS df MS Number of obs = 14
-------------+------------------------------ F( 2, 11) = 997.41
Model | 48033207.2 2 24016603.6 Prob > F = 0.0000
Residual | 264869.664 11 24079.0604 R-squared = 0.9945
-------------+------------------------------ Adj R-squared = 0.9935
Total | 48298076.8 13 3715236.68 Root MSE = 155.17
------------------------------------------------------------------------------
y | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
x1 | .0409783 .0026847 15.26 0.000 .0350693 .0468874
x2 | 5.142659 .670864 7.67 0.000 3.666098 6.619221
_cons | -3489.548 265.1268 -13.16 0.000 -4073.088 -2906.008
------------------------------------------------------------------------------
2、根据以上,两种方法结果,最终模型都是要删除 x3,x4,x5 解释变量吗?