18 Aug
2011
18 Aug
'11
4:46 a.m.
Dear Everyone,
I have attempted to run some multinomial logit models with mixed
results with the Zelig package.
I very much baffled with the fact that what ran about two weeks ago on
the same computer, is not working any longer. Neither are course
examples that also ran on the computer. The only alteration that
happened was I downloaded R Studio, although was NOT using it to run my
models in the end.
I would appreciate if anyone has an idea what went wrong...!
Thank you and kind regards,
Zsofia Ignacz
Complete syntax:
baseyear <- zelig(as.factor(chair) ~ v5+v8+v67+fHU+fCZ+fEG+fw32+fw33,
model="mlogit", data=isjp2)
summary (baseyear)
hypHU91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=1, fCZ=0, fEG=0,
fw32=1, fw33=0)
hypCZ91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=0, fCZ=1, fEG=0,
fw32=0, fw33=0)
hypEG91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=0, fCZ=0, fEG=1,
fw32=0, fw33=0)
hypWG91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=0, fCZ=0, fEG=0,
fw32=0, fw33=0)
hypprobHU91 <- sim(baseyear, x=hypHU91 , num=1500)
###Error message: "Error in UseMethod("cov") : no applicable method for
'vcov' applied to an object of class "c('vglm', 'vlm',
'vlmsmall')"
--
Zsófia Ignácz
PhD Student
Berlin Graduate School of Social Sciences
European PhD for Social-economic and Social Statistics
Institut für Sozialwissenschaften
Humboldt Universität zu Berlin
Unter den Linden 6
D 10099 Berlin
Office: Luisenstrasse 56, Room 208, 10177 Berlin
zsofia.ignacz(a)sowi.hu-berlin.de
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
19 Aug
19 Aug
10:35 a.m.
Dear Zelig Developers,
The problem that I described in my previous email seems to be resolved
only by reinstalling R completely.
After reinstalling R, the exact same script with sim (), which didn't
run before, runs.
I also reproduced the "vcov" error message on a different computer and
again, the remedy we reinstalling R completely.
Thus, I highly suspect there is a bug somewhere either in the package
or in R itself. I have already submitted a bug report to the R
developers.
I would be grateful if the question will be resolved,
Kind regards,
Zsofia Ignacz
On Thu, 18 Aug 2011 10:46:29 +0200, Zsófia Ignácz wrote:
Dear Everyone,
I have attempted to run some multinomial logit models with mixed
results with the Zelig package.
I very much baffled with the fact that what ran about two weeks ago
on the same computer, is not working any longer. Neither are course
examples that also ran on the computer. The only alteration that
happened was I downloaded R Studio, although was NOT using it to run
my models in the end.
I would appreciate if anyone has an idea what went wrong...!
Thank you and kind regards,
Zsofia Ignacz
Complete syntax:
baseyear <- zelig(as.factor(chair) ~ v5+v8+v67+fHU+fCZ+fEG+fw32+fw33,
model="mlogit", data=isjp2)
summary (baseyear)
hypHU91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=1, fCZ=0, fEG=0,
fw32=1, fw33=0)
hypCZ91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=0, fCZ=1, fEG=0,
fw32=0, fw33=0)
hypEG91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=0, fCZ=0, fEG=1,
fw32=0, fw33=0)
hypWG91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=0, fCZ=0, fEG=0,
fw32=0, fw33=0)
hypprobHU91 <- sim(baseyear, x=hypHU91 , num=1500)
###Error message: "Error in UseMethod("cov") : no applicable method
for 'vcov' applied to an object of class "c('vglm',
'vlm',
'vlmsmall')"
--
Zsófia Ignácz
PhD Student
Berlin Graduate School of Social Sciences
European PhD for Social-economic and Social Statistics
Institut für Sozialwissenschaften
Humboldt Universität zu Berlin
Unter den Linden 6
D 10099 Berlin
Office: Luisenstrasse 56, Room 208, 10177 Berlin
zsofia.ignacz(a)sowi.hu-berlin.de
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
--
Zsófia Ignácz
Berlin Graduate School of Social Sciences
European PhD for Social-economic and Social Statistics
Institut für Sozialwissenschaften
Humboldt Universität zu Berlin
Unter den Linden 6
D 10099 Berlin
Office: Luisenstrasse 56, Room 208, 10177 Berlin
zsofia.ignacz(a)sowi.hu-berlin.de
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
10:40 a.m.
Well, to be precise, it sounds like the problem is with RStudio.
R and Zelig work together quite nicely.
Best, Olivia
On Fri, Aug 19, 2011 at 10:35 AM, Zsófia Ignácz <ignaczzs(a)hu-berlin.de>wrote;wrote:
Dear Zelig Developers,
The problem that I described in my previous email seems to be resolved only
by reinstalling R completely.
After reinstalling R, the exact same script with sim (), which didn't run
before, runs.
I also reproduced the "vcov" error message on a different computer and
again, the remedy we reinstalling R completely.
Thus, I highly suspect there is a bug somewhere either in the package or in
R itself. I have already submitted a bug report to the R developers.
I would be grateful if the question will be resolved,
Kind regards,
Zsofia Ignacz
On Thu, 18 Aug 2011 10:46:29 +0200, Zsófia Ignácz wrote:
Dear Everyone,
I have attempted to run some multinomial logit models with mixed
results with the Zelig package.
I very much baffled with the fact that what ran about two weeks ago
on the same computer, is not working any longer. Neither are course
examples that also ran on the computer. The only alteration that
happened was I downloaded R Studio, although was NOT using it to run
my models in the end.
I would appreciate if anyone has an idea what went wrong...!
Thank you and kind regards,
Zsofia Ignacz
Complete syntax:
baseyear <- zelig(as.factor(chair) ~ v5+v8+v67+fHU+fCZ+fEG+fw32+**fw33,
model="mlogit", data=isjp2)
summary (baseyear)
hypHU91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=1, fCZ=0, fEG=0,
fw32=1, fw33=0)
hypCZ91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=0, fCZ=1, fEG=0,
fw32=0, fw33=0)
hypEG91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=0, fCZ=0, fEG=1,
fw32=0, fw33=0)
hypWG91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=0, fCZ=0, fEG=0,
fw32=0, fw33=0)
hypprobHU91 <- sim(baseyear, x=hypHU91 , num=1500)
###Error message: "Error in UseMethod("cov") : no applicable method
for 'vcov' applied to an object of class "c('vglm',
'vlm',
'vlmsmall')"
--
Zsófia Ignácz
PhD Student
Berlin Graduate School of Social Sciences
European PhD for Social-economic and Social Statistics
Institut für Sozialwissenschaften
Humboldt Universität zu Berlin
Unter den Linden 6
D 10099 Berlin
Office: Luisenstrasse 56, Room 208, 10177 Berlin
zsofia.ignacz(a)sowi.hu-berlin.**de <zsofia.ignacz(a)sowi.hu-berlin.de>
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options:
http://lists.gking.harvard.**edu/?info=zelig<http://lists.gking.harvard.…
Zelig program information:
http://gking.harvard.edu/**zelig/<http://gking.harvard.edu/zelig/>
--
Zsófia Ignácz
Berlin Graduate School of Social Sciences
European PhD for Social-economic and Social Statistics
Institut für Sozialwissenschaften
Humboldt Universität zu Berlin
Unter den Linden 6
D 10099 Berlin
Office: Luisenstrasse 56, Room 208, 10177 Berlin
zsofia.ignacz(a)sowi.hu-berlin.**de <zsofia.ignacz(a)sowi.hu-berlin.de>
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options:
http://lists.gking.harvard.**edu/?info=zelig<http://lists.gking.harvard.…
Zelig program information:
http://gking.harvard.edu/**zelig/<http://gking.harvard.edu/zelig/>
11:13 a.m.
It turns out that the problem was in library incompatibilities. Zsófia was
loading 20 packages at the beginning of the R session, with Zelig somewhere
in the middle. Every time you load a library in R, the functions in that
library overwrite the functions of the same name in base R and previous R
libraries that you have loaded.
Suppose that there is a vcov function in the car library, and you load car
after you load Zelig. Then R will try to use the vcov function from car when
Zelig asks for a vcov function, and Zelig will not work.
Therefore, you should only load the libraries that you need at the beginning
of your R session, and you should always load Zelig last. Zelig will load
additional libraries automatically as it needs them. If you find that Zelig
is bombing when it had previously worked, quit your R session, restart, load
just the Zelig package, and start again.
There is no need to continually reinstall
Best,
Olivia
On Thu, Aug 18, 2011 at 4:46 AM, Zsófia Ignácz <ignaczzs(a)hu-berlin.de>wrote;wrote:
Dear Everyone,
I have attempted to run some multinomial logit models with mixed results
with the Zelig package.
I very much baffled with the fact that what ran about two weeks ago on the
same computer, is not working any longer. Neither are course examples that
also ran on the computer. The only alteration that happened was I downloaded
R Studio, although was NOT using it to run my models in the end.
I would appreciate if anyone has an idea what went wrong...!
Thank you and kind regards,
Zsofia Ignacz
Complete syntax:
baseyear <- zelig(as.factor(chair) ~ v5+v8+v67+fHU+fCZ+fEG+fw32+**fw33,
model="mlogit", data=isjp2)
summary (baseyear)
hypHU91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=1, fCZ=0, fEG=0, fw32=1,
fw33=0)
hypCZ91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=0, fCZ=1, fEG=0, fw32=0,
fw33=0)
hypEG91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=0, fCZ=0, fEG=1, fw32=0,
fw33=0)
hypWG91 <- setx(baseyear, v5=1, v8=50, v67=4, fHU=0, fCZ=0, fEG=0, fw32=0,
fw33=0)
hypprobHU91 <- sim(baseyear, x=hypHU91 , num=1500)
###Error message: "Error in UseMethod("cov") : no applicable method for
'vcov' applied to an object of class "c('vglm', 'vlm',
'vlmsmall')"
--
Zsófia Ignácz
PhD Student
Berlin Graduate School of Social Sciences
European PhD for Social-economic and Social Statistics
Institut für Sozialwissenschaften
Humboldt Universität zu Berlin
Unter den Linden 6
D 10099 Berlin
Office: Luisenstrasse 56, Room 208, 10177 Berlin
zsofia.ignacz(a)sowi.hu-berlin.**de <zsofia.ignacz(a)sowi.hu-berlin.de>
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options:
http://lists.gking.harvard.**edu/?info=zelig<http://lists.gking.harvard.…
Zelig program information:
http://gking.harvard.edu/**zelig/<http://gking.harvard.edu/zelig/>
4657
days inactive
4658
days old
3 comments
2 participants
participants (2)
-
Olivia Lau -
Zsófia Ignácz