vif共线性诊断_52stata

个人资料

微博

正文字体大小：大中小

vif共线性诊断

(2013-10-15 21:21:26)

标签：

分类： STATA与统计学

方差膨胀因子（Variance Inflation Factor，VIF）：容忍度的倒数，VIF越大，显示共线性越严重。

经验判断方法表明：当0<VIF<10，不存在多重共线性；当10≤VIF<100，存在较强的多重共线性；当

VIF≥100，存在严重多重共线性

***

. reg age abl1 pab1 glu1

Source | SS df MS Number of obs = 326

-------------+------------------------------ F( 3, 322) = 6.47

Model | 6113.76336 3 2037.92112 Prob > F = 0.0003

Residual | 101469.246 322 315.121881 R-squared = 0.0568

-------------+------------------------------ Adj R-squared = 0.0480

Total | 107583.009 325 331.024644 Root MSE = 17.752

------------------------------------------------------------------------------

age | Coef. Std. Err. t P>|t| [95% Conf. Interval]

-------------+----------------------------------------------------------------

abl1 | -.0468193 .1789086 -0.26 0.794 -.3987967 .3051581

pab1 | -.0625732 .0231875 -2.70 0.007 -.1081913 -.016955

glu1 | .4274023 .2663446 1.60 0.110 -.096593 .9513977

_cons | 57.09872 6.284078 9.09 0.000 44.73569 69.46176

------------------------------------------------------------------------------

. vif

Variable | VIF 1/VIF

-------------+----------------------

abl1 | 1.82 0.547978

pab1 | 1.67 0.600082

glu1 | 1.14 0.879441

-------------+----------------------

Mean VIF | 1.54

阅读┊ 收藏 ┊ 喜欢 ▼ ┊打印┊举报/Report