Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: st: different results of estat vif in different versions
From
Yang Jue <[email protected]>
To
Stata <[email protected]>
Subject
RE: st: different results of estat vif in different versions
Date
Wed, 2 Mar 2011 08:14:13 +0000
Thank you very much for your mail. Here are my results.
1) Result by STATA 11
Linear regression Number of obs = 416 F( 25, 151) = 4.57 Prob > F = 0.0000 R-squared = 0.1368 Root MSE = .46207
(Std. Err. adjusted for 152 clusters in id)------------------------------------------------------------------------------ | Robust unfccc | Coef. Std. Err. t P>|t| [95% Conf. Interval]-------------+---------------------------------------------------------------- logpopsize | .0215847 .0206976 1.04 0.299 -.0193097 .062479 logimfgdppc | .0448861 .0312921 1.43 0.154 -.0169409 .106713growthrate~e | -.0010476 .0035075 -0.30 0.766 -.0079777 .0058824 freedom | .0804955 .0373796 2.15 0.033 .006641 .1543501 engonumber | -.0063351 .0032835 -1.93 0.056 -.0128226 .0001524 iengobranch | .0272889 .0276632 0.99 0.325 -.027368 .0819457 proportion | .0091545 .1436593 0.06 0.949 -.2746873 .2929963ruralpopul~n | .0020349 .0016403 1.24 0.217 -.0012061 .0052759 forestarea | -.!
0008872 .0008976 -0.99 0.325 -.0026608 .0008863 eumember | .000867 .117996 0.01 0.994 -.2322693 .2340034Industryin~e | -.0022502 .0024063 -0.94 0.351 -.0070045 .0025042 igonumber | -.0006447 .0014637 -0.44 0.660 -.0035367 .0022472wtreatynor~g | -.1292887 .1311666 -0.99 0.326 -.3884475 .1298701 wlanlag | .1209351 .0945647 1.28 0.203 -.0659057 .307776 wrellag | -.0680624 .0995165 -0.68 0.495 -.2646871 .1285622 wgnipclag | .1021731 .0796936 1.28 0.202 -.0552854 .2596316 wdistance | .0169681 .0335349 0.51 0.614 -.0492901 .0832263 year1994 | .2815527 .0695474 4.05 0.000 .1441411 .4189644 year1995 | .3551343 .1259585 2.82 0.005 .1062657 .6040029 year1996 | .2183863 .1655258 1.32 0.189 -.1086594 .545432 year1997 | .1105864 .1783407 0.62 0.536 -.2417789 !
.4629518 year1998 | .1882554 .2079546 0.91 0.367 -.222
6211 .5991319 year1999 | .0976277 .2187639 0.45 0.656 -.3346057 .5298611 year2000 | .6075577 .2958538 2.05 0.042 .0230102 1.192105 year2004 | .4269236 .4233205 1.01 0.315 -.4094725 1.26332 _cons | -.1347726 .2652201 -0.51 0.612 -.6587942 .3892491------------------------------------------------------------------------------
. . estat vif
Variable | VIF 1/VIF -------------+---------------------- logimfgdppc | 4.99 0.200312wtreatynor~g | 3.57 0.279870 iengobranch | 3.29 0.304173 proportion | 2.96 0.337535ruralpopul~n | 2.71 0.368402 logpopsize | 2.38 0.420121 wdistance | 2.12 0.471930 igonumber | 2.02 0.494521 eumember | 1.99 0.501583 freedom | 1.84 0.543222Industryin~e | 1.66 0.603218 engonumber | 1.62 0.617358 wrellag | 1.48 0.674094 year1994 | 1.43 0.697412 wlanlag | 1.23 0.809883 wgnipclag | 1.22 0.821512 forestarea | 1.18 0.844083growthrate~e | 1.13 0.887723 year1995 | 1.12 0.889045 year1999 | 1.09 0.918495 year2000 | 1.08 0.926113 year1996 | 1.07 0.931968 year2004 | 1.07 0.932725 year1998 | 1.07 0.933773 year1997 | 1.06 0.940307-------------+-!
--------------------- Mean VIF | 1.86
2) Result by STATA 10
. estat vif
Variable | VIF 1/VIF -------------+----------------------
wtreatynor~g | 7.01 0.142566 logimfgdppc | 5.11 0.195883 wdistance | 4.95 0.201820 year1995 | 3.31 0.302183ruralpopul~n | 3.02 0.331124 wrellag | 2.90 0.344992 year1996 | 2.84 0.351572 iengobranch | 2.81 0.356165 proportion | 2.69 0.371232 logpopsize | 2.40 0.416472 year1997 | 2.36 0.423087 year1998 | 2.29 0.436870 igonumber | 2.28 0.438634Industryin~e | 2.09 0.478245 engonumber | 2.01 0.498266 eumember | 1.93 0.519186 year1999 | 1.88 0.532898 year1994 | 1.88 0.533076 wlanlag | 1.87 0.533476 year2000 | 1.78 0.560443 freedom | 1.71 0.584250 wgnipclag | 1.47 0.682361 forestarea | 1.31 0.763710growthrate~e | 1.25 0.797988 year2004 | 1.17 0.856524 -------------+---------------------- Mean VIF | 2.57
Best regards,Jue Yang
Center for Global Environmental ResearchNational Institute for Environmental Studies, Japan
----------------------------------------
> Date: Tue, 1 Mar 2011 23:47:36 -0800
> From: [email protected]
> To: [email protected]
> Subject: Re: st: different results of estat vif in different versions
>
> Greetings
> Perhaps you could post your results from the two versions to show the differences. I
> created an example with some highly collinear variables which would seem to create
> challenges in computing VIFs. The results from my example look very much the same in the
> two versions....
>
> ---- Version 10 ----
>
> . which estat
> C:\Program Files\Stata10\ado\base\e\estat.ado
> *! version 1.0.0 03jan2005
>
> . sysuse auto, clear
> (1978 Automobile Data)
> . gen mm = mpg*mpg
> . gen mmm = mpg*mpg*mpg
> . quietly regress price rep78 headroom trunk weight turn displacement gear_ratio ///
> foreign mpg mm mmm
> . estat vif
>
> Variable | VIF 1/VIF
> -------------+----------------------
> mm | 3642.36 0.000275
> mmm | 973.82 0.001027
> mpg | 947.38 0.001056
> weight | 12.89 0.077585
> displacement | 12.01 0.083293
> turn | 5.07 0.197348
> gear_ratio | 4.64 0.215490
> foreign | 3.52 0.284084
> trunk | 2.77 0.361185
> rep78 | 2.08 0.481183
> headroom | 1.86 0.538475
> -------------+----------------------
> Mean VIF | 509.85
>
> --- Version 11.1 -----------
>
> . which estat
> C:\Program Files\Stata11\ado\base\e\estat.ado
> *! version 1.0.1 20mar2009
>
> . sysuse auto, clear
> (1978 Automobile Data)
> . gen mm = mpg*mpg
> . gen mmm = mpg*mpg*mpg
> . quietly regress price rep78 headroom trunk weight turn displacement gear_ratio ///
> foreign mpg mm mmm
> . estat vif
>
> Variable | VIF 1/VIF
> -------------+----------------------
> mm | 3642.36 0.000275
> mmm | 973.82 0.001027
> mpg | 947.38 0.001056
> weight | 12.89 0.077585
> displacement | 12.01 0.083293
> turn | 5.07 0.197348
> gear_ratio | 4.64 0.215490
> foreign | 3.52 0.284084
> trunk | 2.77 0.361185
> rep78 | 2.08 0.481183
> headroom | 1.86 0.538475
> -------------+----------------------
> Mean VIF | 509.85
>
> Best regards,
>
> Michael N. Mitchell
> Data Management Using Stata - http://www.stata.com/bookstore/dmus.html
> A Visual Guide to Stata Graphics - http://www.stata.com/bookstore/vgsg.html
> Stata tidbit of the week - http://www.MichaelNormanMitchell.com
>
>
>
> On 2011-03-01 11.25 PM, Yang Jue wrote:
> >
> > Dear all, I am running a simple regress model both in STATA 10 and STATA 11. The results of the model (also Coef. and R2) are complete the same, but the results of estat vif are different.
> > I checked the release STATA 11 to see if they changed the command 'estat vif' in different versions. But I cannot find out.
> > Are there any differences between these two versions?
> > Thank you very much for your help.
> > Yang
> > Center for Global Environmental ResearchNational Institute for Environmental Studies, Japan
> >
> >
> > *
> > * For searches and help try:
> > * http://www.stata.com/help.cgi?search
> > * http://www.stata.com/support/statalist/faq
> > * http://www.ats.ucla.edu/stat/stata/
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/