I spent a bit of time trying to install 2.6.1 on Windows with R 2.2.1,
but this seems to fail because the 2.6.1 package has no DESCRIPTION
file. It installs fine on R 2.3.0. Perhaps the release notes could
mention something?
Here's the failure message I get on 2.2.1:
> install.packages("Zelig", repos = "http://gking.harvard.edu")
trying URL 'http://gking.harvard.edu/bin/windows/contrib/2.2/Zelig_2.6-1.zip'
Content type 'application/zip' length 1303902 bytes
opened URL
downloaded 1273Kb
Error in gzfile(file, "r") : unable to open connection
In addition: Warning message:
cannot open compressed file 'Zelig/DESCRIPTION'
Cheers
David Hugh-Jones
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options: http://lists.hmdc.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
Hi Julia,
Your "euneg" variable does not appear in the matchit() or zelig(). They
have to belong to those formulae in order for setx() to set the value of
the variable.
Hope this helps,
Kosuke
-----------------------------------------------------
Kosuke Imai Office: Corwin Hall 041
Assistant Professor Phone: 609-258-6601
Department of Politics eFax: 973-556-1929
Princeton University Email: kimai(a)Princeton.Edu
Princeton, NJ 08544-1012 http://imai.princeton.edu
-----------------------------------------------------
On Sun, 7 May 2006, Julia Gray wrote:
> Hi, Kosuke. It was just least squares; I'm attaching the whole bit of
> code below.
>
> library(foreign)
> library(zelig)
> library(MatchIt)
>
> eu<-read.dta("may74.dta")
>
>
> m.out0<- matchit(euneg ~ cpid + xfin + reserves + cacct + forexstd +
> weur + ce + baltics + balkans, method = "nearest", data = eu)
>
> z.out0 <- zelig(ytsp3 ~ cpid + xfin + reserves + cacct + forexstd +
> weur + ce + baltics + balkans, data = match.data(m.out0), model="ls")
>
> x.out0 <- setx(z.out0, euneg=0)
>
> x.out1 <- setx(z.out0, euneg=1)
> s.out0 <- sim(z.out0, x = x.out0)
> summary(s.out0)
>
>
>
>
> z.out2 <- zelig(mkt1 ~ lgdp + linf + cacct + debtsrv + latin + europe +
> africa, data = match.data(m.out0, "treat"), model="ls")
>
> x.out2 <- setx(z.out2, data = match.data(m.out0, "control"), cond = TRUE)
> s.out2 <- sim(z.out2, x = x.out2)
>
> ate.all <- c(s.out0$qi$ate.ev, -s.out2$qi$ate.ev)
>
> mean(ate.all)
> sd(ate.all)
> quantile(ate.all, c(0.025, 0.975))
>
> Quoting Kosuke Imai <kimai(a)Princeton.Edu>:
>
> > Dear Julia,
> > Could you let us know what model you were running with zelig()?
> > Thanks,
> > Kosuke
> >
> > On Sun, 7 May 2006, Julia Gray wrote:
> >
> >> Hi,
> >>
> >> I'm running into trouble with the model-based estimates. After setting
> >> the IVs at their means
> >> > x.out0 <- setx(z.out0, myvar=0)
> >> >
> >> > x.out1 <- setx(z.out0, myvar=1)
> >>
> >>
> >> all goes well up to this point:
> >>
> >> > s.out0 <- sim(z.out1, x = x.out1)
> >>
> >> at which point I get the following error:
> >> > Error in any(class(object) == "MCMCZelig") :
> >> object "z.out1" not found
> >>
> >> Running through the example as it was defined above, it appears as
> >> though z.out1 wasn't actually defined, though z.out0 was. But trying to
> >> modify the code produces another error:
> >>
> >> > sim(z.out0, x = x.out0)
> >> > Error in mvrnorm(num, mu = coef(object), Sigma = vcov.lm(object)) :
> >> incompatible arguments
> >>
> >> Incidentally, I ran matchit a few months ago on a different dataset
> >> with no problems, but now when I try to use the same code, I get the
> >> same error message listed directly above.
> >>
> >> Any help you might offer would be greatly appreciated.
> >>
> >> best, Julia
> >>
> >> -
> >> MatchIt mailing list served by Harvard-MIT Data Center
> >> List Address: matchit(a)latte.harvard.edu
> >> Subscribe/Unsubscribe: http://lists.hmdc.harvard.edu/?info=matchit
> >> MatchIt Software and Documentation: http://gking.harvard.edu/matchit/
> >>
> >
> >
>
>
>
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options: http://lists.hmdc.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
If matching involve numerous covariates and thus the QQ plots are two or
more, is there a syntax that will plot all QQ plots in a single window or
on a page? Or do I need to play with lattice?
Any help would be greatly appreciated.
Ani
Anirudh V. S. Ruhil, Ph.D.
Sr. Research Associate
Voinovich Center for Leadership and Public Affairs
Ohio University
Building 21, The Ridges
Athens, OH 45701-2979
Tel: 740.597.1949 | Fax: 740.597.3057
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options: http://lists.hmdc.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
Dear Julia,
Could you let us know what model you were running with zelig()?
Thanks,
Kosuke
On Sun, 7 May 2006, Julia Gray wrote:
> Hi,
>
> I'm running into trouble with the model-based estimates. After setting
> the IVs at their means
> > x.out0 <- setx(z.out0, myvar=0)
> >
> > x.out1 <- setx(z.out0, myvar=1)
>
>
> all goes well up to this point:
>
> > s.out0 <- sim(z.out1, x = x.out1)
>
> at which point I get the following error:
> > Error in any(class(object) == "MCMCZelig") :
> object "z.out1" not found
>
> Running through the example as it was defined above, it appears as
> though z.out1 wasn't actually defined, though z.out0 was. But trying to
> modify the code produces another error:
>
> > sim(z.out0, x = x.out0)
> > Error in mvrnorm(num, mu = coef(object), Sigma = vcov.lm(object)) :
> incompatible arguments
>
> Incidentally, I ran matchit a few months ago on a different dataset
> with no problems, but now when I try to use the same code, I get the
> same error message listed directly above.
>
> Any help you might offer would be greatly appreciated.
>
> best, Julia
>
> -
> MatchIt mailing list served by Harvard-MIT Data Center
> List Address: matchit(a)latte.harvard.edu
> Subscribe/Unsubscribe: http://lists.hmdc.harvard.edu/?info=matchit
> MatchIt Software and Documentation: http://gking.harvard.edu/matchit/
>
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options: http://lists.hmdc.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
Hi All,
I'd like to created graphic depictions of interval estimates for the mean
of a group of attributes compared across two samples. So, for example, x1 x2
and x3 are questions answered in two surveys and, rather than doing
frequentist t-tests of their differences, I'd like to produce Bayesian HPD's
of either the means themselves (and see where they do or do not overlap) or
produce Bayesian HPD's of the differences.
I was wondering if anyone had experience doing this sort of thing in Zelig
and/or could point me to an example.
Many thanks,
Joe Retzer