Re: [R] VIF threshold implying multicollinearity
Quite, but apparently not a boisterous one? John Kane Kingston ON Canada -Original Message- From: cfly...@ncsu.edu Sent: Mon, 27 Jul 2015 14:35:06 -0400 To: jrkrid...@inbox.com Subject: Re: [R] VIF threshold implying multicollinearity No actually it is a quiet good paper! :) On Mon, Jul 27, 2015 at 8:14 AM, John Kane jrkrid...@inbox.com wrote: +1 I, originally, read it as a stringent criticism of the first paper. John Kane Kingston ON Canada -Original Message- From: r.tur...@auckland.ac.nz Sent: Mon, 27 Jul 2015 15:12:43 +1200 To: cfly...@ncsu.edu Subject: Re: [R] VIF threshold implying multicollinearity On 27/07/15 13:36, Collin Lynch wrote: The following sources discuss the issues generally and may be a goof pointer to the literature ... SNIP I think that the foregoing merits fortune status! :-) cheers, Rolf -- Technical Editor ANZJS Department of Statistics University of Auckland Phone: +64-9-373-7599 ext. 88276 __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help [https://stat.ethz.ch/mailman/listinfo/r-help] PLEASE do read the posting guide http://www.R-project.org/posting-guide.html [http://www.R-project.org/posting-guide.html] and provide commented, minimal, self-contained, reproducible code. FREE ONLINE PHOTOSHARING - Share your photos online with your friends and family! Visit http://www.inbox.com/photosharing [http://www.inbox.com/photosharing] to find out more! FREE 3D MARINE AQUARIUM SCREENSAVER - Watch dolphins, sharks orcas on your desktop! __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] VIF threshold implying multicollinearity
+1 I, originally, read it as a stringent criticism of the first paper. John Kane Kingston ON Canada -Original Message- From: r.tur...@auckland.ac.nz Sent: Mon, 27 Jul 2015 15:12:43 +1200 To: cfly...@ncsu.edu Subject: Re: [R] VIF threshold implying multicollinearity On 27/07/15 13:36, Collin Lynch wrote: The following sources discuss the issues generally and may be a goof pointer to the literature ... SNIP I think that the foregoing merits fortune status! :-) cheers, Rolf -- Technical Editor ANZJS Department of Statistics University of Auckland Phone: +64-9-373-7599 ext. 88276 __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. FREE ONLINE PHOTOSHARING - Share your photos online with your friends and family! Visit http://www.inbox.com/photosharing to find out more! __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] VIF threshold implying multicollinearity
No actually it is a quiet good paper! :) On Mon, Jul 27, 2015 at 8:14 AM, John Kane jrkrid...@inbox.com wrote: +1 I, originally, read it as a stringent criticism of the first paper. John Kane Kingston ON Canada -Original Message- From: r.tur...@auckland.ac.nz Sent: Mon, 27 Jul 2015 15:12:43 +1200 To: cfly...@ncsu.edu Subject: Re: [R] VIF threshold implying multicollinearity On 27/07/15 13:36, Collin Lynch wrote: The following sources discuss the issues generally and may be a goof pointer to the literature ... SNIP I think that the foregoing merits fortune status! :-) cheers, Rolf -- Technical Editor ANZJS Department of Statistics University of Auckland Phone: +64-9-373-7599 ext. 88276 __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. FREE ONLINE PHOTOSHARING - Share your photos online with your friends and family! Visit http://www.inbox.com/photosharing to find out more! [[alternative HTML version deleted]] __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] VIF threshold implying multicollinearity
Dear All I have a general question about VIF. While there are multiple rules of thumb about the threshold value of VIF, e.g. 4 or 10, implying multicollinearity, I am wondering if anyone can point me to some literature supporting these rules of thumb. Thank you so much! wensui __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] VIF threshold implying multicollinearity
On 27/07/15 13:36, Collin Lynch wrote: The following sources discuss the issues generally and may be a goof pointer to the literature ... SNIP I think that the foregoing merits fortune status! :-) cheers, Rolf -- Technical Editor ANZJS Department of Statistics University of Auckland Phone: +64-9-373-7599 ext. 88276 __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] VIF threshold implying multicollinearity
The following sources discuss the issues generally and may be a goof
pointer to the literature on VIF. Particularly the Schroeder paper.
@article{Yi:Evaluation,
AUTHOR = {Youjae Yi},
TITLE = {On the Evaluation of Main Effects in Multiplicative
Regression Models.},
JOURNAL = {Journal of the Market Research Society},
VOLUME = {31},
NUMBER = {1},
MONTH = {January},
YEAR= {1989},
PAGES = {133-138}
}
@article{Gordon:Issues,
AUTHOR = {Robert A. Gordon},
TITLE = {Issues in Multiple Regression},
JOURNAL = {American Journal of Sociology},
VOLUME = {73},
NUMBER = {5},
MONTH = {March},
YEAR= {1968},
PAGES = {592-616}
}
@misc{Lynch:Multicollinearity,
author = {Scott M. Lynch},
title = {Multicollinearity},
year = {2003},
url= {\url{
http://www.princeton.edu/~slynch/soc504/multicollinearity.pdf}},
note = [Online; accessed 11-October-2013]
}
@article{Schroeder:Multicollinearity,
AUTHOR = {Mary Ann Schroeder
and Janice Lander
and Stacey Levine-Silverman},
TITLE = {Diagnosing and Dealing with Multicollinearity},
JOURNAL = {Western Journal of Nursing Research},
VOLUME = {12},
NUMBER = {2},
YEAR= {1990},
PAGES = {175-187}
}
@book{Afifi:Computer,
AUTHOR= {A. Afifi and V. Clark},
TITLE = {Computer-aided Multivariate Analysis},
PUBLISHER = {Wadsworth, Belmont California},
YEAR = {1984}
}
On Sun, Jul 26, 2015 at 5:00 PM, Wensui Liu liuwen...@gmail.com wrote:
Dear All
I have a general question about VIF.
While there are multiple rules of thumb about the threshold value of
VIF, e.g. 4 or 10, implying multicollinearity, I am wondering if
anyone can point me to some literature supporting these rules of
thumb.
Thank you so much!
wensui
__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[[alternative HTML version deleted]]
__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] vif in package car: there are aliased coefficients in the model
Of Rodolfo Pelinson Sent: March-27-15 3:07 PM To: r-help@r-project.org Subject: [R] vif in package car: there are aliased coefficients in the model Hello. I'm trying to use the function vif from package car in a lm. However it returns the following error: Error in vif.default(lm(MDescores.sitescores ~ hidroperiodo + localizacao + : there are aliased coefficients in the model When I exclude any predictor from the model, it returns this warning message: Warning message: In cov2cor(v) : diag(.) had 0 or NA entries; non-finite result is doubtful When I exclude any other predictor from the model vif finally works. I can't figure it out whats the problem. This are the results that R returns me: vif(lm(MDescores.sitescores ~ hidroperiodo + localizacao + area + profundidade + NTVM + NTVI + PCs...c.1.., data = MDVIF)) Error in vif.default(lm(MDescores.sitescores ~ hidroperiodo + localizacao + : there are aliased coefficients in the model vif(lm(MDescores.sitescores ~ localizacao + area + profundidade + NTVM + NTVI + PCs...c.1.., data = MDVIF)) GVIF Df GVIF^(1/(2*Df)) localizacao NaN 2 NaN area NaN 1 NaN profundidade NaN 1 NaN NTVM NaN 1 NaN NTVI NaN 1 NaN PCs...c.1.. NaN 1 NaN Warning message: In cov2cor(v) : diag(.) had 0 or NA entries; non-finite result is doubtful Thanks. -- Rodolfo Mei Pelinson. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting- guide.html and provide commented, minimal, self-contained, reproducible code. --- This email has been checked for viruses by Avast antivirus software. http://www.avast.com -- Rodolfo Mei Pelinson. John Fox, Professor McMaster University Hamilton, Ontario, Canada http://socserv.mcmaster.ca/jfox/ -- Rodolfo Mei Pelinson. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] vif in package car: there are aliased coefficients in the model
Dear Rodolfo, Sending the data helps, though if you had done what I suggested, you would have seen what's going on: snip -- dim(data) [1] 8 8 summary(lm(response_variable ~ predictor_1 + predictor_2 + predictor_3 + predictor_4 + + predictor_5 + predictor_6 + predictor_7, data = data)) Call: lm(formula = response_variable ~ predictor_1 + predictor_2 + predictor_3 + predictor_4 + predictor_5 + predictor_6 + predictor_7, data = data) Residuals: ALL 8 residuals are 0: no residual degrees of freedom! Coefficients: (1 not defined because of singularities) Estimate Std. Error t value Pr(|t|) (Intercept) -5.1905 NA NA NA predictor_1yellow 2.4477 NA NA NA predictor_2fora 6.5056 NA NA NA predictor_2interior 6.0769 NA NA NA predictor_3 0.6750 NA NA NA predictor_4 3.0742 NA NA NA predictor_5 0.6715 NA NA NA predictor_6 -0.9850 NA NA NA predictor_7 NA NA NA NA Residual standard error: NaN on 0 degrees of freedom Multiple R-squared: 1, Adjusted R-squared:NaN F-statistic: NaN on 7 and 0 DF, p-value: NA snip -- So the data set that you're using has 8 cases and 8 variables, one of which is a factor with 3 levels. Consequently, the model you're fitting my LS has 9 coefficients. Necessarily the rank of the model matrix is deficient. When you eliminate a coefficient, you get a perfect fit: 8 coefficients fit to 8 cases with 0 df for error. This is of course nonsense: You don't have enough data to fit a model of this complexity. In fact, you might not have enough data to reasonably fit a model with just 1 predictor. I'm cc'ing this response to the r-help email list, where you started this thread. Best, John On Sat, 28 Mar 2015 12:04:05 -0300 Rodolfo Pelinson rodolfopelin...@gmail.com wrote: Thanks a lot for your answer and your time! But Im still having the same problem. That's the script I am using: library(car) data -read.table(data_vif.txt, header = T, sep = \t, row.names = 1) data vif(lm(response_variable ~ predictor_1 + predictor_2 + predictor_3 + predictor_4 + predictor_5 + predictor_6 + predictor_7, data = data)) vif(lm(response_variable ~ predictor_1 + predictor_2 + predictor_3 + predictor_4 + predictor_5 + predictor_6, data = data)) the first vif function above returns me the following error: Error in vif.default(lm(response_variable ~ predictor_1 + predictor_2 + : there are aliased coefficients in the model Then if I remove any one of the predictors (in the script I removed predictor_7 as an example), it returns this: GVIF Df GVIF^(1/(2*Df)) predictor_1 NaN 1 NaN predictor_2 NaN 2 NaN predictor_3 NaN 1 NaN predictor_4 NaN 1 NaN predictor_5 NaN 1 NaN predictor_6 NaN 1 NaN Warning message: In cov2cor(v) : diag(.) had 0 or NA entries; non-finite result is doubtful Can you help me with this? I even attached to this e-mail my data set. It's a small table. Sorry for the question. 2015-03-27 21:51 GMT-03:00 John Fox j...@mcmaster.ca: Dear Rodolfo, It's apparently the case that at least one of the columns of the model matrix for your model is perfectly collinear with others. There's not nearly enough information here to figure out exactly what the problem is, and the information that you provided certainly falls short of allowing me or anyone else to reproduce your problem and diagnose it properly. It's not even clear from your message exactly what the structure of the model is, although localizacao is apparently a factor with 3 levels. If you look at the summary() output for your model or just print it, you should at least see which coefficients are aliased, and that might help you understand what went wrong. I hope this helps, John --- John Fox, Professor McMaster University Hamilton, Ontario, Canada http://socserv.mcmaster.ca/jfox/ -Original Message- From: R-help [mailto:r-help-boun...@r-project.org] On Behalf Of Rodolfo Pelinson Sent: March-27-15 3:07 PM To: r-help@r-project.org Subject: [R] vif in package car: there are aliased coefficients in the model Hello. I'm trying to use the function vif from package car in a lm. However it returns the following error: Error in vif.default(lm
[R] vif in package car: there are aliased coefficients in the model
Hello. I'm trying to use the function vif from package car in a lm. However it returns the following error: Error in vif.default(lm(MDescores.sitescores ~ hidroperiodo + localizacao + : there are aliased coefficients in the model When I exclude any predictor from the model, it returns this warning message: Warning message: In cov2cor(v) : diag(.) had 0 or NA entries; non-finite result is doubtful When I exclude any other predictor from the model vif finally works. I can't figure it out whats the problem. This are the results that R returns me: vif(lm(MDescores.sitescores ~ hidroperiodo + localizacao + area + profundidade + NTVM + NTVI + PCs...c.1.., data = MDVIF)) Error in vif.default(lm(MDescores.sitescores ~ hidroperiodo + localizacao + : there are aliased coefficients in the model vif(lm(MDescores.sitescores ~ localizacao + area + profundidade + NTVM + NTVI + PCs...c.1.., data = MDVIF)) GVIF Df GVIF^(1/(2*Df)) localizacao NaN 2 NaN area NaN 1 NaN profundidade NaN 1 NaN NTVM NaN 1 NaN NTVI NaN 1 NaN PCs...c.1.. NaN 1 NaN Warning message: In cov2cor(v) : diag(.) had 0 or NA entries; non-finite result is doubtful Thanks. -- Rodolfo Mei Pelinson. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] vif in package car: there are aliased coefficients in the model
Dear Rodolfo, It's apparently the case that at least one of the columns of the model matrix for your model is perfectly collinear with others. There's not nearly enough information here to figure out exactly what the problem is, and the information that you provided certainly falls short of allowing me or anyone else to reproduce your problem and diagnose it properly. It's not even clear from your message exactly what the structure of the model is, although localizacao is apparently a factor with 3 levels. If you look at the summary() output for your model or just print it, you should at least see which coefficients are aliased, and that might help you understand what went wrong. I hope this helps, John --- John Fox, Professor McMaster University Hamilton, Ontario, Canada http://socserv.mcmaster.ca/jfox/ -Original Message- From: R-help [mailto:r-help-boun...@r-project.org] On Behalf Of Rodolfo Pelinson Sent: March-27-15 3:07 PM To: r-help@r-project.org Subject: [R] vif in package car: there are aliased coefficients in the model Hello. I'm trying to use the function vif from package car in a lm. However it returns the following error: Error in vif.default(lm(MDescores.sitescores ~ hidroperiodo + localizacao + : there are aliased coefficients in the model When I exclude any predictor from the model, it returns this warning message: Warning message: In cov2cor(v) : diag(.) had 0 or NA entries; non-finite result is doubtful When I exclude any other predictor from the model vif finally works. I can't figure it out whats the problem. This are the results that R returns me: vif(lm(MDescores.sitescores ~ hidroperiodo + localizacao + area + profundidade + NTVM + NTVI + PCs...c.1.., data = MDVIF)) Error in vif.default(lm(MDescores.sitescores ~ hidroperiodo + localizacao + : there are aliased coefficients in the model vif(lm(MDescores.sitescores ~ localizacao + area + profundidade + NTVM + NTVI + PCs...c.1.., data = MDVIF)) GVIF Df GVIF^(1/(2*Df)) localizacao NaN 2 NaN area NaN 1 NaN profundidade NaN 1 NaN NTVM NaN 1 NaN NTVI NaN 1 NaN PCs...c.1.. NaN 1 NaN Warning message: In cov2cor(v) : diag(.) had 0 or NA entries; non-finite result is doubtful Thanks. -- Rodolfo Mei Pelinson. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting- guide.html and provide commented, minimal, self-contained, reproducible code. --- This email has been checked for viruses by Avast antivirus software. http://www.avast.com __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] vif
Dear Arun,Thanks indeed
Date: Thu, 3 Oct 2013 10:22:38 -0700
From: smartpink...@yahoo.com
To: r-help@r-project.org
Subject: Re: [R] vif
Hi Eliza,
Then, res needs a slight modification
library(car)
res- lapply(colnames(h),function(x) {x1- h[,x];dat1-
do.call(rbind,lapply(seq_len(ncol(mat1)),function(i){ x2- m[,mat1[,i]];GG-
lm(x1~x2[,1]+x2[,2]+x2[,3]+x2[,4]);GGsum- summary(GG); data.frame(
Models=paste(colnames(x2),collapse=,), Multiple_Rsq= GGsum$r.squared,
Adjusted_Rsq = GGsum$adj.r.squared, Pval =
paste(GGsum$coef[-1,4],collapse=,),Vif=paste(vif(GG),collapse=,),stringsAsFactors=FALSE)
})); dat1[rev(order(dat1[,3])),][1:10,]})
names(res)- colnames(h)
A.K.
From: eliza botto eliza_bo...@hotmail.com
To: smartpink...@yahoo.com smartpink...@yahoo.com
Sent: Thursday, October 3, 2013 12:42 PM
Subject: vif
Dear Arun,
There is one small question however.
what if i also want in the table a column for vif values of each model.
vif values can be generated for any model in the following way
GG-lm(h[,any column]~m[,any column]+m[,any other column] +m[,any other
column] +m[,any other column])
library(car)
vif(GG)
Here will be get 4 vif values. I want to make a new column which could
contain these values seperated by comma, very much similar to the way we did
with Pr(|t|) values.
thanks in advance
elisa
Date: Thu, 3 Oct 2013 09:01:53 -0700
From: smartpink...@yahoo.com
To: r-help@r-project.org
Subject: Re: [R] a simple question
Hi,
Try:
set.seed(494)
h- matrix(sample(1:40,4*124,replace=TRUE),ncol=4)
set.seed(39)
m- matrix(sample(1:100,10*124,replace=TRUE),ncol=10)
colnames(h)- paste0(h,1:4)
colnames(m)- paste0(m,1:10)
mat1-combn(colnames(m),4)
res- lapply(colnames(h),function(x) {x1- h[,x];dat1-
do.call(rbind,lapply(seq_len(ncol(mat1)),function(i){ x2-
m[,mat1[,i]];GG- lm(x1~x2[,1]+x2[,2]+x2[,3]+x2[,4]);GGsum- summary(GG);
data.frame( Models=paste(colnames(x2),collapse=,), Multiple_Rsq=
GGsum$r.squared, Adjusted_Rsq = GGsum$adj.r.squared, Pval =
paste(GGsum$coef[-1,4],collapse=,),stringsAsFactors=FALSE) }));
dat1[rev(order(dat1[,3])),][1:10,]})
names(res)- colnames(h)
A.K.
From: eliza botto eliza_bo...@hotmail.com
To: smartpink...@yahoo.com smartpink...@yahoo.com
Sent: Thursday, October 3, 2013 11:07 AM
Subject: a simple question
Dear Arun,
I hope you are fine. I actually
wanted to discuss the following problem.
I have a linear model of the
following form.
GG-lm(h[,any column]~m[,any
column]+m[,any other column] +m[,any other column] +m[,any other column])
where,
h is matrix with 4 columns and
124 rows
m is matrix with 10 columns and
124 rows
what I want is the following
make a loop command to run the
linear model of all the possible combinations of columns of m with each
column of h.
more precisely, if i take column
1 of matrix h, it should be linear modeled with every combination of 10
(210
combinations) columns of m.
All the columns of h m
have certain names (you can suppose any). The summary(GG) will give
Multiple R-squared,Adjusted R-squared and 4 values of Pr(|t|).
I want in the end a table in the
following format.
Models
Multiple R-squaredAdjusted R-squared Pr(|t|)
Name of columns of m separated by
comma Multiple R-squared Adjusted R-squared Pr(|t|)
separated by comma
For Example
Models
Multiple R-squaredAdjusted R-squared Pr(|t|)
eliza, allen, murphy, jack
0.544 0.56
0.000114,0.000112,0.01114,0.002114
where,
eliza, allen, murphy, jack are column names.
The models are to be enlisted in the order of their Adjusted R-squared
values. The models with highest Adjusted R-squared value should be on the
top and so on. i m only interested in top 10 models. so the remaining
should be ignored.
I tried to put in my question everything but if there is anything wrong plz
inform me.
Thankyou very much in advance,
Eliza
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide
Re: [R] vif
Hi Eliza,
Then, res needs a slight modification
library(car)
res- lapply(colnames(h),function(x) {x1- h[,x];dat1-
do.call(rbind,lapply(seq_len(ncol(mat1)),function(i){ x2- m[,mat1[,i]];GG-
lm(x1~x2[,1]+x2[,2]+x2[,3]+x2[,4]);GGsum- summary(GG); data.frame(
Models=paste(colnames(x2),collapse=,), Multiple_Rsq= GGsum$r.squared,
Adjusted_Rsq = GGsum$adj.r.squared, Pval =
paste(GGsum$coef[-1,4],collapse=,),Vif=paste(vif(GG),collapse=,),stringsAsFactors=FALSE)
})); dat1[rev(order(dat1[,3])),][1:10,]})
names(res)- colnames(h)
A.K.
From: eliza botto eliza_bo...@hotmail.com
To: smartpink...@yahoo.com smartpink...@yahoo.com
Sent: Thursday, October 3, 2013 12:42 PM
Subject: vif
Dear Arun,
There is one small question however.
what if i also want in the table a column for vif values of each model.
vif values can be generated for any model in the following way
GG-lm(h[,any column]~m[,any column]+m[,any other column] +m[,any other column]
+m[,any other column])
library(car)
vif(GG)
Here will be get 4 vif values. I want to make a new column which could
contain these values seperated by comma, very much similar to the way we did
with Pr(|t|) values.
thanks in advance
elisa
Date: Thu, 3 Oct 2013 09:01:53 -0700
From: smartpink...@yahoo.com
To: r-help@r-project.org
Subject: Re: [R] a simple question
Hi,
Try:
set.seed(494)
h- matrix(sample(1:40,4*124,replace=TRUE),ncol=4)
set.seed(39)
m- matrix(sample(1:100,10*124,replace=TRUE),ncol=10)
colnames(h)- paste0(h,1:4)
colnames(m)- paste0(m,1:10)
mat1-combn(colnames(m),4)
res- lapply(colnames(h),function(x) {x1- h[,x];dat1-
do.call(rbind,lapply(seq_len(ncol(mat1)),function(i){ x2- m[,mat1[,i]];GG-
lm(x1~x2[,1]+x2[,2]+x2[,3]+x2[,4]);GGsum- summary(GG); data.frame(
Models=paste(colnames(x2),collapse=,), Multiple_Rsq= GGsum$r.squared,
Adjusted_Rsq = GGsum$adj.r.squared, Pval =
paste(GGsum$coef[-1,4],collapse=,),stringsAsFactors=FALSE) }));
dat1[rev(order(dat1[,3])),][1:10,]})
names(res)- colnames(h)
A.K.
From: eliza botto eliza_bo...@hotmail.com
To: smartpink...@yahoo.com smartpink...@yahoo.com
Sent: Thursday, October 3, 2013 11:07 AM
Subject: a simple question
Dear Arun,
I hope you are fine. I actually
wanted to discuss the following problem.
I have a linear model of the
following form.
GG-lm(h[,any column]~m[,any
column]+m[,any other column] +m[,any other column] +m[,any other column])
where,
h is matrix with 4 columns and
124 rows
m is matrix with 10 columns and
124 rows
what I want is the following
make a loop command to run the
linear model of all the possible combinations of columns of “m” with each
column of “h”.
more precisely, if i take column
1 of matrix “h”, it should be linear modeled with every combination of 10 (210
combinations) columns of “m”.
All the columns of “h” “m”
have certain names (you can suppose any). The summary(GG) will give Multiple
R-squared, Adjusted R-squared and 4 values of Pr(|t|).
I want in the end a table in the
following format.
Models
Multiple R-squared Adjusted R-squared Pr(|t|)
Name of columns of m separated by
comma Multiple R-squared Adjusted R-squared Pr(|t|)
separated by comma
For Example
Models
Multiple R-squared Adjusted R-squared Pr(|t|)
eliza, allen, murphy, jack
0.544 0.56
0.000114,0.000112,0.01114,0.002114
where,
eliza, allen, murphy, jack are column names.
The models are to be enlisted in the order of their Adjusted R-squared
values. The models with highest Adjusted R-squared value should be on the top
and so on. i m only interested in top 10 models. so the remaining should be
ignored.
I tried to put in my question everything but if there is anything wrong plz
inform me.
Thankyou very much in advance,
Eliza
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] vif calculation with car and HH packages
Dear all, I have faced a problem while calculating VIF values via the packages, car and HH for the models witout intercepts. Below is an illustrative example: 1) via the car package y-rnorm(100,0,1) x1-rnorm(100,0,1) x2-rnorm(100,0,1) x3-rnorm(100,0,1) model1-lm(y~-1+x1+x2+x3) model2-lm(y~-1+x1+x2) library(car) vif(model1) x1 x2 x3 1.000279 1.019231 1.019376 Warning message: In vif.lm(model1) : No intercept: vifs may not be sensible. vif(model2) x1 x2 1.85 1.85 Warning message: In vif.lm(model2) : No intercept: vifs may not be sensible. 2) via the HH package library(HH) vif(model1) x2 x3 1.000557 1.000557 vif(model2) Error in vif.default(xx, na.action = na.action) : vif requires two or more X-variables. I could not understand why this occured. Does anyone have any idea about it? Best Ozgur - Ozgur ASAR Research Assistant Middle East Technical University Department of Statistics 06531, Ankara Turkey Ph: 90-312-2105309 -- View this message in context: http://r.789695.n4.nabble.com/vif-calculation-with-car-and-HH-packages-tp4555402p4555402.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] vif calculation with car and HH packages
Dear Özgür, car::vif() produces a warning, not an error. It will proceed to compute VIFs based on the correlation matrix of the coefficients (take a look at car:::vif.lm) even if there is no intercept, and even though this would not normally correspond to variance inflation due to correlation of the predictors. If you think that makes sense, then by all means use the VIFs. Best, John John Fox Sen. William McMaster Prof. of Social Statistics Department of Sociology McMaster University Hamilton, Ontario, Canada http://socserv.mcmaster.ca/jfox/ On Fri, 13 Apr 2012 09:49:54 -0700 (PDT) Özgür Asar oa...@metu.edu.tr wrote: Dear all, I have faced a problem while calculating VIF values via the packages, car and HH for the models witout intercepts. Below is an illustrative example: 1) via the car package y-rnorm(100,0,1) x1-rnorm(100,0,1) x2-rnorm(100,0,1) x3-rnorm(100,0,1) model1-lm(y~-1+x1+x2+x3) model2-lm(y~-1+x1+x2) library(car) vif(model1) x1 x2 x3 1.000279 1.019231 1.019376 Warning message: In vif.lm(model1) : No intercept: vifs may not be sensible. vif(model2) x1 x2 1.85 1.85 Warning message: In vif.lm(model2) : No intercept: vifs may not be sensible. 2) via the HH package library(HH) vif(model1) x2 x3 1.000557 1.000557 vif(model2) Error in vif.default(xx, na.action = na.action) : vif requires two or more X-variables. I could not understand why this occured. Does anyone have any idea about it? Best Ozgur - Ozgur ASAR Research Assistant Middle East Technical University Department of Statistics 06531, Ankara Turkey Ph: 90-312-2105309 -- View this message in context: http://r.789695.n4.nabble.com/vif-calculation-with-car-and-HH-packages-tp4555402p4555402.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] vif calculation with car and HH packages
Dear Prof. Fox, I got the point, things are clear now. Thank you very much, Best wishes Ozgur - Ozgur ASAR Research Assistant Middle East Technical University Department of Statistics 06531, Ankara Turkey Ph: 90-312-2105309 -- View this message in context: http://r.789695.n4.nabble.com/vif-calculation-with-car-and-HH-packages-tp4555402p4555653.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] vif function using lm object
On Wed, Dec 21, 2011 at 9:28 AM, arunkumar akpbond...@gmail.com wrote: Hi, can anyone please explain why the vif should have more than 2 terms. *vif.lm(lmobj) : model contains fewer than 2 terms* why it is throwng error if it is one variable. The _VIF_ is a measure of _multicollinearity_, which occurs between two or more predictors in a regression model. In other words, you cannot have collinearity in a single predictor. Also, to compute the VIF you need to regress one regressor on all the remaining regressors. If you have a single regressor, then you cannot estimate a VIF. See the related articles on Wikipedia. Regards Liviu - Thanks in Advance Arun -- View this message in context: http://r.789695.n4.nabble.com/vif-function-using-lm-object-tp4220904p4220904.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Do you know how to read? http://www.alienetworks.com/srtest.cfm http://goodies.xfce.org/projects/applications/xfce4-dict#speed-reader Do you know how to write? http://garbl.home.comcast.net/~garbl/stylemanual/e.htm#e-mail __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] vif function using lm object
Hi, can anyone please explain why the vif should have more than 2 terms. *vif.lm(lmobj) : model contains fewer than 2 terms* why it is throwng error if it is one variable. - Thanks in Advance Arun -- View this message in context: http://r.789695.n4.nabble.com/vif-function-using-lm-object-tp4220904p4220904.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] VIF for logit models
Hello,
I wonder whether it's possible to use vif{car} for binary logit models
(estimated by using glm() function). And what about a case if all
explanatory variables are binary as well? Is VIF still a good criterion
for multicollinearity detection?
Thanks a lot.
Tomas
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
