In discussing Garrett's paper, "King Tomz and WIttenberg" finds that
left labor power has a distinguishable effect only when exposure to trade or
capital mobility is high. The paper finds Garrett's claim that "left" always
outspends the "right", regardless of level of market integration, could be
right, as a tendency-propensity, but it could be also attributed to chance
alone.
In light of my new found awareness of Post-treatment Bias, and subject to
Garrett's data set(the countries he looked at), could we look as Garrett
being correct because, richer countries(where the "left" when in power can
spend more), tend to have a more "open" market-economy? The propensity to
spend more could always be there for the "left" regardless, but in the
absence of resources, how can they spend?
Of course if the findings take the country variable as fixed (same country
or countries) and are based on specific countries pulling a switch from
"closed" to "open" (since we are dealing with panel data), my thinking is
flawed. "King Tomz and WIttenberg" though, quotes Garrett as "...where
countries were highly integrated into the international economy ...",
implying that Garrett distinguishes among countries... which allows me to
think the possibility exists...
Agelos
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Hello 2001istas,
I was wondering if anyone had encountered this error message before
when using Zelig or the ordered logit function in R:
"In function (formula, data, weights, start, ..., subset, na.action, :
design appears to be rank-deficient, so dropping some coefs"
I'm getting it in response to an ordered logit model I'm trying to
fit, specifically when I try to put in a factor variable for year
dummies. What does it mean/what should I do? The regression runs but
drops the year dummy factor variable. I get the same problem if I
create my own dummy variables for the years, too.
Thanks for your help!
--Alex
--
Alexander Hertel-Fernandez
ahertel(a)fas.harvard.edu
alexander.hertel(a)gmail.com
765.430.2063
_______________________________________________
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
Gov 2001,
We had an excellent suggestion this week from a student on the reading
quiz. She suggested that it might be nice for true R novices to have
a space to help each other out with really basic R questions without
having to bombard the list. Now generally, we think if you have a
question, you should absolutely send it to the list (or search the
list archives!). Its the fastest way to get answers.
However, under the theory that some people will hold back their
questions rather than send them to the list, we have created the
R-Help Forum (on the course website). Feel free to post here about
any R related questions. Molly and I will check in occasionally and
experienced R users can get "participation credit" by showing up to
share their expertise.
Hopefully some of you will find this useful. That being said, keep
using the list!
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
Hi Folks,
For those doing the challenge problem, there was a typo in my original
slides including the EM update step for Y*. I changed it (Friday
morning, I think), but I think some people are still referencing the
old, erroneous material. Sorry!
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
Yes, that works. Actually, I'm now wondering if the problem was in using the coding at the bottom of pg. 2, because it never pulled up the data under the variable name with "variablename". But the datasetname$variablename works.
Thanks!
Laurel
From: Adam Chilton [mailto:adam.s.chilton@gmail.com]
Sent: Sunday, March 06, 2011 1:02 PM
To: Laurel Grassin-Drake
Subject: Re: [gov2001] data from STATA
Hi Laurel,
I spent forever trying to load the data and get the variables right, so I know how frustrating that can be. My guess is the problem you are having is that you are not converting it to be a data frame. You are welcome to take a look at my code for loading it though.
library(foreign)
data <- read.dta("fearondata.dta")
as.data.frame(data)
dim(data)
names(data)
head(data)
vars.m1 <- data.frame(data$onset, data$warl , data$gdpenl, data$lpopl1, data$lmtnest, data$ncontig, data$Oil, data$nwstate, data$instab, data$polity2l, data$ethfrac, data$relfrac)
data2 <- na.omit(vars.m1)
which(data2[,1]==4)
data2[2396,1] <- 1
data2[2396,1]
head(data2)
as.data.frame(data2)
On Sun, Mar 6, 2011 at 12:54 PM, Laurel Grassin-Drake <lgrassin(a)mit.edu<mailto:lgrassin@mit.edu>> wrote:
Thanks for the help.
But now I've been struggling with trying to use the data. I've finally figured out that R seems to be separating the data into groups such that it is creating a matrix with fewer columns and more rows. So it takes the first X variables and populates n rows with the data, but the next group of variables start at row n+1 instead of column X+1.
Anyone know what I need to do?
Thank you.
Laurel
From: gov2001-l-bounces(a)lists.fas.harvard.edu<mailto:gov2001-l-bounces@lists.fas.harvard.edu> [mailto:gov2001-l-bounces@lists.fas.harvard.edu<mailto:gov2001-l-bounces@lists.fas.harvard.edu>] On Behalf Of Colin Sullivan
Sent: Sunday, March 06, 2011 9:54 AM
To: Class List for Gov 2001/E-2001
Subject: Re: [gov2001] co-authors
Laurel - you have to use the "foreign" library; then it will find the command.
Colin
On Sun, Mar 6, 2011 at 9:42 AM, Laurel Grassin-Drake <lgrassin(a)mit.edu<mailto:lgrassin@mit.edu>> wrote:
Dear all,
I know this is pretty basic, but how do I import the data set for this week's pset into R? I've tried read.dta, but the program tells me that it couldn't find the function.
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<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
Hi everyone,
Just a reminder that the deadline to choosing a paper to replicate and
having it approved is this Wednesday, March 2. If you are still looking for
a co-author, please e-mail me (not the Class list) and include a brief
description of what subject areas you are interested in. I will help you
find a co-author. Extension school students, you can ignore this e-mail.
Hope everyone is having a great weekend!
Best,
Molly
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Is anyone else having difficulty accessing last years lectures? It's never
been a problem for me before, but it doesn't seem to be working at the
moment - it seems to be getting stuck trying to load the list of all the
lectures (the same thing is happening in both Firefox and Safari so doesn't
seem to be a browser issue)
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Hi class,
Maybe someone can answer a question I and my partner have about our
replication paper. It says modeling children's schooling decisions "results
in the standard linear-in-means empirical model." Is that just a fancy way
of saying that children's schooling decisions can be modeled using linear
regression?
Thanks!
Adela
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Gov 2001,
It sounds like some people are still having trouble accessing the
section/lecture videos from this year. If you are still having a
problem, we're sorry! I am sending another request to our IT contact
so I want to get the info for everyone who is still having trouble.
If you can't access the video please send me:
1. Your Harvard ID #
2. Your School and University (FAS, Harvard for example would be mine)
Thanks everyone! Sorry this is still an issue.
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