Hi everyone,
Zelig version 3.4-0 is online (see http://gking.harvard.edu/zelig)
In this version we introduce a new scheme to cite zelig models.
- zelig() function now takes an extra argument, "citation". If "citation" is true (default)
the model citation is printed out.
- (for contributors) describe.mymodel function now requires two more attributes:
authors: a vector which should contain the authors who contributed to "mymodel" to Zelig.
year: the year when "mymodel" was added to Zelig
- Our building script generate the model citation on the fly, so there is no need for the contributors
to include any citation in the model documentation.
There are some bug fixes too:
- Fixed the incompatibility of Zelig with the new version of lme4 package. _Note that there is still
a problem with simulation step of "gamma.mixed" model. The contributor is still working on that.
- Fixed the bug with "gam" models (wrong predicted values)
- Fixed the bug when zelig model name was provided from a variable (reported from Jeroen)
Please try it out and let us know. Any feedback before we submit the package to cran would be much
appreciated. Yes, this release is tested against R-2.8.0
As usually, to update/install this release:
- *nix / windows users:
install.packages("Zelig", repos="http://gking.harvard.edu" <http://gking.harvard.edu%22>)
- mac users:
install.packages("Zelig", repos="http://gking.harvard.edu"
<http://gking.harvard.edu%22>, type="source")
Please let us know if you run into problems.
thanks,
Ferdi
We will be releasing a new version *very* soon!
Kosuke
--
Department of Politics
Princeton University
http://imai.princeton.edu
On Mon, 27 Oct 2008, Maritza Dowling wrote:
> Hello!
> I am using R (2.8). I am getting the following message. Could you tell me
> what I am missing?
>
> z.out <- zelig(re78 ~ treat + age + educ + black + hispan + nodegree +
> married + re74 + re75, model = "ls" , data = match.data(m.out))
> Error in zelig.default(re78 ~ treat + age + educ + black + hispan + nodegree
> + :
> no slot "call" in the result
>
> THANK YOU A LOT!
>
> NMDowling
>
>
>
-
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/
Dear Zelig team,
I have a question about the plot.surv() function. Since it plots
Kaplan-Meier survival curves, does it work after estimating models with
time-varying covariates?
Thank you,
Jose Aleman, Ph.D.
Assistant Professor
Political Science Department
Fordham University
-
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/
I just updated R to version 2.8.0, but Zelig seems to have big issues with
this version of R. Even very simple models return errors:
> data(macro)
> z.out1 <- zelig(unem ~ gdp + capmob + trade, model = "normal",data =
macro)
Error in zelig.default(unem ~ gdp + capmob + trade, model = "normal", :
no slot "call" in the result
Same problems occur on win32 and linux. Should I try to downgrade R, or will
there be a quick fixing release for Zelig in a couple of days?
Jeroen
Can you tell me your syntax? I'm not sure how you are using plot.zelig()
to plot the propensity score. Are you using the plot function that comes
with MatchIt?
If the issue is just about the title, then what you can do is to go into
the pdf file of your plot (that is, open it in a text editor such as
emacs, NotePad, WinEdt, etc.), then do search and find "Distribution of
Propensity Scores" and simply delete those words (keep the quotation
marks).
Best,
Kosuke
--
Department of Politics
Princeton University
http://imai.princeton.edu
On Sun, 19 Oct 2008, Beaujean, Alex wrote:
> Drs. Imai & King:
>
> Hi. I am using your MatchIt and Zelig programs, and plotted
> (type="jitter") the propensity scores for matched participants via the
> plot.zelig() function. The title: "Distribution of Propensity Scores"
> seems to be hardcoded into this output. I am wondering if you knew how
> to turn this off. The journal I'm submitting the manuscript to doesn't
> want titles on the graphics.
>
> Thanks for any help you can offer,
>
> Alex
> -*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-
> A. Alexander Beaujean, PhD
> Licensed Psychologist
> Assistant Professor, Department of Educational Psychology
> Director, Baylor Psychometric Laboratory
> One Bear Place #97301
> Waco, TX 76798-7301
> Alex_Beaujean(a)Baylor.edu
> Ph: 254.710.1548
> Fax: (254) 710-3265
> http://myprofile.cos.com/abeaujean
> http://homepages.baylor.edu/psychometric_lab
>
> "I'm always amazed to see a psychologist offering a glib explanation of
> some immensely complex behavioral incident when psychological science
> has not even provided explanations for comparatively simple general
> phenomena..." --Arthur R. Jensen
>
> "At one time psychology courses were the recognized refuge of those who
> wished to take a 'science' without facing the rigors of mathematics--as
> encountered, for example, in chemistry and physics. Nowadays the
> psychologist, and even the biologist need to be as good at mathematics
> as the engineering student." --Raymond B. Cattell
>
>
-
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/
Dear Zelig developers,
Today I had the idea to extend Zelig through Rcmdr, which would
involve creating an Rcmdr plug-in. I'm still new to R, and certainly
new to Zelig, but at a first glance this seems trivial to me.
I would be interested in developing such a plug-in, but first would
like to know your opinion on such an enterprise. Would it be
reasonable to develop in the first place? How would you imagine a
similar plug-in? Would there be any Zelig-related "gotchas"? If I am
right in my perception, Zelig together with Rcmdr could effectively
become the software of choice for non-statisticians and
non-programmers.
Regards,
Liviu
-
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/