Gov 2001 Students,
For those who are still having access issues to this year's lecture videos
and section videos, we have a new procedure. Hopefully the last round fixed
the problems, but in case it didn't- the DCE people sent the following
instructions:
In regards to the authentication problems, can you ask students to submit
there issues to fill:
de-bugreport(a)dce.harvard.edu
Things they need to include when submitting their problem.
1) A detailed a description of what happens when they try to login,
including any error messages that may appear.
2) Browser version
3) Flash version (can be found by going to
http://kb2.adobe.com/cps/155/tn_15507.html)
4) OS (Mac/Win) and version (Mac 10.5, Win XP, Win 7, etc.)
5) Type of connection (cable, DSL) and if they're using wireless
Note that this is NOT for the Video annotation tool. if you are having
problems with that tool (or NB or anything else) send an email to me.
Thank you everyone for your continued patience with this. Have a great
spring break!
Brandon
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Hi everyone,
Julie and I have a quick question for some code we're trying to run in R.
How can we cluster standard errors for a logit in Zelig? In stata its just
logit Y X1 X2 X3, cluster(variable), where the variable in cluster() is a
dummy variable for each survey round we're using, but I'm not sure how to
make the equivalent syntax in R. Does anyone know this off the top of their
head?
Thanks!
- Noah
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
A reminder. This is a really good conference, informal, intense. You'll
meet lots of people. Worth doing.
Gary King
http://gking.harvard.edu
617-500-7570
(from my phone)
---------- Forwarded message ----------
From: "Jeff Gill" <jgill(a)wustl.edu>
Date: Mar 9, 2011 11:07 AM
Subject: [POLMETH] Polmeth Deadline Looming
To: <POLMETH(a)artsci.wustl.edu>
The deadline for applying to attend/present at the 28th Annual Summer
Meeting of the Society for Political Methodology at Princeton University,
July 28-30, 2011 is March 15 (next Tuesday). Interested scholars should
apply as soon as possible at:
http://polmeth.wustl.edu/conferences/2011/1_apply/apply.php
Keep in mind that March 15 is a HARD deadline and there will be NO
exceptions, unlike many other meetings in political science. The
application process is completely online and only takes minutes.
-Jeff
.............................................................................
Jeff Gill Director......Center for Applied Statistics
Professor.....Department of Political Science
Professor.....Department of Mathematics (by courtesy)
Coursemaster..Clinical Research Training Center, School of Medicine
President.....Society for Political Methodology, APSA Section
..............One Brookings Dr., Seigle Hall, LL-085, Box 1203
..............Washington University, St. Louis, MO 63130-4899
..............Email: jgill(a)wustl.edu, Phone: 314-935-9012
..............URL: jgill.wustl.edu, Fax: 314-935-4141
.............................................................................
**********************************************************
Political Methodology E-Mail List
Editors: Diana O'Brien <dzobrien(a)wustl.edu>
Jon C. Rogowski <rogowski.jon(a)gmail.com>
**********************************************************
Send messages to polmeth(a)artsci.wustl.edu
To join the list, cancel your subscription, or modify
your subscription settings visit:
http://polmeth.wustl.edu/polmeth.php
**********************************************************
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Gov 2001,
There was some confusion, but I wanted to remind everyone that we will
have section tonight (Spring Break hasn't started yet :))
Brandon
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
--
gov2001 mailing list served by Harvard-MIT Data Center
List Address: gov2001(a)lists.gking.harvard.edu
Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=gov2001
> For the GDP plotting (2.3), I am confused about the variable - gdpenl. It
> does not go from 250-8000$. Am I using the wrong variable? Or do I have to
> transform it first?
Thanks
Slawa
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Hi everyone,
I've received a couple e-mails saying people are getting this error when
they are running their optimization function:
> y<-conflictdata$onset
> X<-subset(conflictdata, select=c(warl, gdpenl, lpopl1, lmtnest, ncontig,
Oil, nwstate, instab, polity2l, ethfrac, relfrac))
> prob1<-optim(par=rep(0,12), fn=ll.logit, y=y, X=X, method="BFGS",
control=list(fnscale=-1, maxit=1000), hessian=TRUE)
Error in X %*% beta : requires numeric/complex matrix/vector arguments
Usually what is happening here is that your X or your beta are not of type
"matrix". Often when you define X as a subset of the data as it is here, it
comes up as a dataframe rather than a matrix, and R won't matrix multiply
dataframes. The solution here is to use:
X <- as.matrix(X)
and if needed
beta <- as.matrix(beta)
The command: class(X) is really useful because it will tell you what class
or type X is, and then you can change it if it's not a matrix. Also, you'll
get a similar error message when the dimensions of X or beta are such that
you can't multiply them together. Using the command: dim(X) to check your
dimensions can be useful in this case.
Hope this helps!
Best,
Molly
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Thanks, Noah!!
L
From: Noah Nathan [mailto:noah.nathan@gmail.com]
Sent: Tuesday, March 08, 2011 1:41 PM
To: Laurel Grassin-Drake
Subject: Re: [gov2001] zelig rare
Hi Laurel,
The variable you have as "polity21" is actually "polity2l" (2 lower case L). It should run when you change that.
- Noah
On Tue, Mar 8, 2011 at 1:35 PM, Laurel Grassin-Drake <lgrassin(a)mit.edu<mailto:lgrassin@mit.edu>> wrote:
When I try to run the rare function with the code specified in the problem set, I get the following complaint from R:
> rare <- zelig(onset ~ warl + gdpenl + lpopol1 + lmtnest + ncontig +
Oil + nwstate + instab + polity21 + ethfrac + relfrac,
data = m1, model = "relogit")
Error in `[.data.frame`(d, , all.vars(as.expression(formula))) :
undefined columns selected
> rare.coef <- coef(rare); rare.vcov <- vcov(rare)
Error in coef(rare) : object 'rare' not found
m1 is my original cleaned up data frame (the equivalent of "fearon" in the code given in the slides).
Did anyone else run into this problem? I'm guessing that I somehow need to specify the variables differently from just using their names. Any clues?
Thanks.
Laurel
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu<mailto:gov2001-l@lists.fas.harvard.edu>
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
When I try to run the rare function with the code specified in the problem set, I get the following complaint from R:
> rare <- zelig(onset ~ warl + gdpenl + lpopol1 + lmtnest + ncontig +
Oil + nwstate + instab + polity21 + ethfrac + relfrac,
data = m1, model = "relogit")
Error in `[.data.frame`(d, , all.vars(as.expression(formula))) :
undefined columns selected
> rare.coef <- coef(rare); rare.vcov <- vcov(rare)
Error in coef(rare) : object 'rare' not found
m1 is my original cleaned up data frame (the equivalent of "fearon" in the code given in the slides).
Did anyone else run into this problem? I'm guessing that I somehow need to specify the variables differently from just using their names. Any clues?
Thanks.
Laurel
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Hi everyone,
Two quick things:
First, some of you have been getting answers just slightly off of the Fearon
and Laitin paper (attached) -- some of your answers have been off at the
thousandth decimal place. It turns out depending on the optimization method
you use, you can get just slightly different answers, and Fearon and Laitin
weren't using BFGS in optim. So if a couple of your coefficients are just
slightly off (as in at the 1000th decimal place), don't worry about it.
Second, we are planning on having a class outing at some point and we are
thinking about going mountain biking! So we can get a count of people
interested, we just posted a class poll to the website -- please let us
know whether you would join us in a mountain biking excursion. We'd like to
start planning, so if you can fill it out in the next few days, that would
be great.
Have a great night!
Best,
Molly
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Gov 2001,
The code covered today in lecture is posted on the Lecture Notes page
of the course website (for a lack of a better place to put it). There
was basically zero interest in an extra R section, so we won't be
doing that. However, we encourage you to use the list and use the new
R help forum. Also the code covered today analyzes the same dataset
as your problem set, so if you need help with the initial formatting
you can look there!
Brandon
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
--
gov2001 mailing list served by Harvard-MIT Data Center
List Address: gov2001(a)lists.gking.harvard.edu
Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=gov2001
