24 Jan
2008
24 Jan
'08
6:03 p.m.
Hello,I have estimated a logistic regression model with Zelig using the mi
command to combine multiple imputed datasets. I understand that the summary
output is the combined result and that I can also print out individual
results by dataset. However, in calculating predicted probabilities, I am
wondering which coefficients are being used by Zelig - would these be from
the combined results? In addition, when setting the independent variables
to their means (setx command) - would this be some weighted mean of that
variable across those five datasets? Finally, could someone clarify the
difference between the expected values and the predicted values that are
given in the output.
Thank you,
Amber Wichowsky
University of Wisconsin-Madison
24 Jan
24 Jan
9:35 p.m.
Hi Amber,
All coefficients are used. Basically, you simulate the quantities of
interest based on each of the multiply imputed data sets and then combine
them all. This takes into account imputation uncertainty as well as usual
estimation uncertainty for each imputed data set. When setting the
indpendent variables to their means, setx() will calculate the mean using
all imputed data sets. Finally, the expected values are probabilities,
i.e., Pr(Y = 1 | X), whereas the predicted values take either 1 or 0,
i.e., Y.
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 Thu, 24 Jan 2008, Amber Wichowsky wrote:
Hello,I have estimated a logistic regression model
with Zelig using the mi
command to combine multiple imputed datasets. I understand that the summary
output is the combined result and that I can also print out individual
results by dataset. However, in calculating predicted probabilities, I am
wondering which coefficients are being used by Zelig - would these be from
the combined results? In addition, when setting the independent variables
to their means (setx command) - would this be some weighted mean of that
variable across those five datasets? Finally, could someone clarify the
difference between the expected values and the predicted values that are
given in the output.
Thank you,
Amber Wichowsky
University of Wisconsin-Madison
-
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/
12 Mar
12 Mar
2:08 p.m.
Hello,
I am using Zelig to combine 5 imputed datasets in a very simple logistic
regression. I see that I can get fit statistics when I call up the five
separate results, but is there anyway to get a measure of fit for the
combined model?
Thank you,
Amber Wichowsky
University of Wisconsin-Madison
2:14 p.m.
You can average the five separate fit numbers I'd u like. Zelig doesn't do this
automatically. Gary
---
Sent from my phone; please excuse the terse note.
Gary King
http://gking.harvard.edu
-----Original Message-----
From: "Amber Wichowsky" <wichowsky(a)wisc.edu>
Date: Wed, 12 Mar 2008 13:08:26
To:zelig@lists.gking.harvard.edu
Subject: Re: [zelig] predicted probabilities with zelig and imputed data
Hello,
I am using Zelig to combine 5 imputed datasets in a very simple logistic regression. I see
that I can get fit statistics when I call up the five separate results, but is there
anyway to get a measure of fit for the combined model?
Thank you,
Amber Wichowsky
University of Wisconsin-Madison
-
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/
5912
days inactive
5960
days old
3 comments
3 participants
participants (3)
-
Amber Wichowsky -
Gary King -
Kosuke Imai