20 Jun
2010
20 Jun
'10
3:55 p.m.
I'm running a logit model using the zelig package in R:
z.out <- zelig(trade961a ~ age962 + education962 + personal962 + economy962 +
partisan962 + employment962 + union962 + home962 + market962 + race962 + income962,
data=data96, model="logit")
The dependent variable, trade961a, is a dichotomous factor variable. All other variables
are numeric.
I am interested in personal962, which is:
str(personal962)
num [1:1714] 3 3 1 1 1 2 3 2 2
1 ...
so I run the following to simulate the probability of supporting international trade when
personal962 is 1 (better):
x.out <- setx(z.out, personal962=1)
I get the following error message:
#Error in dta[complete.cases(mf), names(dta) %in% vars, drop = FALSE] :
# incorrect number of dimensions
Any ideas on what I'm doing wrong?
Thanks
Abraham M.
-
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/
20 Jun
20 Jun
4:23 p.m.
From your str ouput, it looks like you may have used
attach(data96)
at some point.
I strongly recommend NOT using attach() -- R (and Zelig) will get
confused. Try detach(data96), run your code again, and see if you
still get the error.
Yours,
Olivia
On Sun, Jun 20, 2010 at 3:55 PM, Mathew, Abraham T <amathew(a)ku.edu> wrote:
I'm running a logit model using the zelig package in R:
z.out <- zelig(trade961a ~ age962 + education962 + personal962 + economy962 +
partisan962 + employment962 + union962 + home962 + market962 + race962 + income962,
data=data96, model="logit")
The dependent variable, trade961a, is a dichotomous factor variable. All other variables
are numeric.
I am interested in personal962, which is:
-
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/
str(personal962)
num [1:1714] 3 3 1 1 1 2
3 2 2 1 ...
so I run the following to simulate the probability of supporting international trade when
personal962 is 1 (better):
x.out <- setx(z.out, personal962=1)
I get the following error message:
#Error in dta[complete.cases(mf), names(dta) %in% vars, drop = FALSE] :
# incorrect number of dimensions
Any ideas on what I'm doing wrong?
Thanks
Abraham M.
-
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/
5:12 p.m.
I make sure to never use the attach() function either.
When I import data, I merely assign it to a variable. For example:
data96 <- read.spss("ANES1996.por")
data96$V9601651 (to get a specific variable)
I never used the attach function.
I will be trying to run the probabilities with glm from the 'car' package.
Thanks
Abraham M.
-----Original Message-----
From: monkeykupo(a)gmail.com on behalf of Olivia Lau
Sent: Sun 6/20/2010 3:23 PM
To: Mathew, Abraham T
Cc: zelig(a)lists.gking.harvard.edu
Subject: Re: [zelig] Logistic Regression with Zelig
From your str ouput, it looks like you may have used
attach(data96)
at some point.
I strongly recommend NOT using attach() -- R (and Zelig) will get
confused. Try detach(data96), run your code again, and see if you
still get the error.
Yours,
Olivia
On Sun, Jun 20, 2010 at 3:55 PM, Mathew, Abraham T <amathew(a)ku.edu> wrote:
I'm running a logit model using the zelig package in R:
z.out <- zelig(trade961a ~ age962 + education962 + personal962 + economy962 +
partisan962 + employment962 + union962 + home962 + market962 + race962 + income962,
data=data96, model="logit")
The dependent variable, trade961a, is a dichotomous factor variable. All other variables
are numeric.
I am interested in personal962, which is:
-
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/
str(personal962)
num [1:1714] 3 3 1 1 1 2
3 2 2 1 ...
so I run the following to simulate the probability of supporting international trade when
personal962 is 1 (better):
x.out <- setx(z.out, personal962=1)
I get the following error message:
#Error in dta[complete.cases(mf), names(dta) %in% vars, drop = FALSE] :
# incorrect number of dimensions
Any ideas on what I'm doing wrong?
Thanks
Abraham M.
-
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/
21 Jun
21 Jun
9:05 p.m.
To diagnose the problem, try using smaller models and figure out when the model breaks
down. That way, you will know which variable is really causing the problem.
Kosuke
Department of Politics
Princeton University
http://imai.princeton.edu
On Jun 20, 2010, at 5:12 PM, Mathew, Abraham T wrote:
I make sure to never use the attach() function either.
When I import data, I merely assign it to a variable. For example:
data96 <- read.spss("ANES1996.por")
data96$V9601651 (to get a specific variable)
I never used the attach function.
I will be trying to run the probabilities with glm from the 'car' package.
Thanks
Abraham M.
-----Original Message-----
From: monkeykupo(a)gmail.com on behalf of Olivia Lau
Sent: Sun 6/20/2010 3:23 PM
To: Mathew, Abraham T
Cc: zelig(a)lists.gking.harvard.edu
Subject: Re: [zelig] Logistic Regression with Zelig
From your str ouput, it looks like you may have used
attach(data96)
at some point.
I strongly recommend NOT using attach() -- R (and Zelig) will get
confused. Try detach(data96), run your code again, and see if you
still get the error.
Yours,
Olivia
On Sun, Jun 20, 2010 at 3:55 PM, Mathew, Abraham T <amathew(a)ku.edu> wrote:
-
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'm running a logit model using the zelig package in R:
z.out <- zelig(trade961a ~ age962 + education962 + personal962 + economy962 +
partisan962 + employment962 + union962 + home962 + market962 + race962 + income962,
data=data96, model="logit")
The dependent variable, trade961a, is a dichotomous factor variable. All other variables
are numeric.
I am interested in personal962, which is:
-
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/ str(personal962)
num [1:1714] 3 3 1 1 1 2
3 2 2 1 ...
so I run the following to simulate the probability of supporting international trade when
personal962 is 1 (better):
x.out <- setx(z.out, personal962=1)
I get the following error message:
#Error in dta[complete.cases(mf), names(dta) %in% vars, drop = FALSE] :
# incorrect number of dimensions
Any ideas on what I'm doing wrong?
Thanks
Abraham M.
-
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/
5080
days inactive
5082
days old
3 comments
3 participants
participants (3)
-
Kosuke Imai -
Mathew, Abraham T -
Olivia Lau