More on understanding variance inflation factors (VIFk) - PowerPoint PPT Presentation

1 / 12
About This Presentation
Title:

More on understanding variance inflation factors (VIFk)

Description:

Title: A Broad Overview of Key Statistical Concepts Author: LSimon Last modified by: LSimon Created Date: 9/16/2002 5:15:16 PM Document presentation format – PowerPoint PPT presentation

Number of Views:26
Avg rating:3.0/5.0
Slides: 13
Provided by: LSi62
Learn more at: https://science.psu.edu
Category:

less

Transcript and Presenter's Notes

Title: More on understanding variance inflation factors (VIFk)


1
More on understanding variance inflation factors
(VIFk)
2
Cement example
3
The regression equation is x4 80.4 - 1.05
x2 Predictor Coef SE Coef T
P Constant 80.396 3.777
21.28 0.000 x2 -1.04657 0.07492
-13.97 0.000 S 4.038 R-Sq 94.7
R-Sq(adj) 94.2
The regression equation is x2 75.3 - 0.905
x4 Predictor Coef SE Coef T
P Constant 75.289 2.204
34.16 0.000 x4 -0.90452 0.06475
-13.97 0.000 S 3.754 R-Sq 94.7
R-Sq(adj) 94.2
Pearson correlation of x2 and x4 -0.973
4
Regress y on x2
The regression equation is y 57.4 0.789
x2 Predictor Coef SE Coef T
P Constant 57.424 8.491
6.76 0.000 x2 0.7891 0.1684
4.69 0.001 S 9.077 R-Sq 66.6
R-Sq(adj) 63.6 Analysis of Variance Source
DF SS MS F
P Regression 1 1809.4 1809.4 21.96
0.001 Residual Error 11 906.3 82.4 Total
12 2715.8
5
Regress y on x4
The regression equation is y 118 - 0.738
x4 Predictor Coef SE Coef T
P Constant 117.568 5.262
22.34 0.000 x4 -0.7382 0.1546
-4.77 0.001 S 8.964 R-Sq 67.5
R-Sq(adj) 64.5 Analysis of
Variance Source DF SS MS
F P Regression 1 1831.9
1831.9 22.80 0.001 Residual Error 11
883.9 80.4 Total 12 2715.8
6
Regress y on x2 and x4
The regression equation is y 94.2 0.311 x2 -
0.457 x4 Predictor Coef SE Coef T
P VIF Constant 94.16 56.63
1.66 0.127 x2 0.3109
0.7486 0.42 0.687 18.7 x4
-0.4569 0.6960 -0.66 0.526
18.7 S 9.321 R-Sq 68.0 R-Sq(adj)
61.6 Analysis of Variance Source DF
SS MS F P Regression
2 1846.88 923.44 10.63 0.003 Residual
Error 10 868.88 86.89 Total 12
2715.76
7
Is the variance of b4 inflated by a factor of
18.7?
almost .
8
Is the variance of b2 inflated by a factor of
18.7?
again almost .
9
Variance inflation factor VIFk
The variance inflation factor quantifies how
much the variance of the estimated regression
coefficient is inflated by the existence of
multicollinearity.
The theory
The estimate
10
Variance inflation factor VIFk
To get the theoretical VIF4,
, that Minitab reports,
we need to multiply the ratio of the variance
estimates by
11
Is the variance of b4 inflated by a factor of
18.7?
12
Is the variance of b2 inflated by a factor of
18.7?
Write a Comment
User Comments (0)
About PowerShow.com