I have a question about interaction terms and clarify that I am not sure
about based on my reading of the archive.
I am using Clarify to get first differences and relative risks. For the
purpose of this example, I have 3 variables of interest. Age, Democracy,
and Age*Democracy.
When I use "setx" to get the mean of the variables, I set things up the
way it is recommended in the Clarify manual for interaction terms.
summarize age1, meanonly /* Compute the mean of x1 */
local mage1=`r(mean)' /* Save the mean in a local macro */
summarize dem1, meanonly /* Compute the mean of x2 */
local mdem1=`r(mean)' /* Save the mean in a local macro */
setx AgeDem1 `mage1'*`mdem1' /* Setx to mean(x1)*mean(x2) */
summarize age2, meanonly /* Compute the mean of x1 */
local mage2=`r(mean)' /* Save the mean in a local macro */
summarize dem2, meanonly /* Compute the mean of x2 */
local mdem2=`r(mean)' /* Save the mean in a local macro */
setx AgeDem1 `mage2'*`mdem2' /* Setx to mean(x1)*mean(x2) */
But I am having trouble figuring out what to do when I subsequently
generate first differences. Do I need to do:
simqi, fd(pr) changex(AgeDem1 min max dem1 min max age1 min max)
OR can I just do
simqi, fd(pr) changex(AgeDem1 min max)
Any guidance you could provide would be much appreciated.
Michael Horowitz
*******************************************************************************
Michael Horowitz
78 Kirkland Street, #1
Cambridge, MA 02138
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.hmdc.harvard.edu/?info=clarify
It is not difficult to simulate the conditional logit probabilities by "hand" with stata (though it's difficult to get the output to look as nice). Once you get started you'll soon realize why it is difficult for a general program like Clarify to simulate conditional logit probabilities-- because stata requires the data to be organized in a quite different way compared to other models.
Katherine M. Harris, Ph.D.
Office of Applied Studies
Substance Abuse and Mental Health Services Administration
5600 Fishers Lane, Rm 16-105
Rockville, MD 20857
Phone: 301.443.0747
Fax: 301.443.9847
email: kharris(a)samhsa.gov
>>> Gary King <king(a)harvard.edu> 05/13/04 03:18PM >>>
not any automatic way. This model would need to be added to Clarify. Or
you could simulate the parmaeters 'by hand' and then compute the quantity
of interests from there, which is basically what Clarify does.
Gary
: Gary King, King(a)Harvard.Edu http://GKing.Harvard.Edu :
: Center for Basic Research Direct (617) 495-2027 :
: in the Social Sciences Assistant (617) 495-9271 :
: 34 Kirkland Street, Rm. 2 HU-MIT DC (617) 495-4734 :
: Harvard U, Cambridge, MA 02138 eFax (617) 812-8581 :
On Thu, 13 May 2004, Jose A. Aleman wrote:
> Dear clarify team,
>
> I am trying to use the xtnbreg (conditional fixed effects negative
> binomial) stata command with clarify, but I'm told by the program that this
> model is not supported. The alternative, nbreg (the negative binomial
> model), which is supported, yields very different results. The pattern of
> dispersion in event count data is very different for each unit, and thus I
> am hesitant to use nbreg, a very different model. Is there any way to get
> around this problem under clarify and retain the original model?
>
> Thank you,
>
> Jose Aleman
> PhD Candidate
> Politics Department
> 130 Corwin Hall
> Princeton, NJ 08544
> 609.937.0190
> 609.258.2147
> <http://www.princeton.edu/~jaaleman>
>
>
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.hmdc.harvard.edu/?info=clarify
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.hmdc.harvard.edu/?info=clarify
Dear clarify team,
I am trying to use the xtnbreg (conditional fixed effects negative
binomial) stata command with clarify, but I'm told by the program that this
model is not supported. The alternative, nbreg (the negative binomial
model), which is supported, yields very different results. The pattern of
dispersion in event count data is very different for each unit, and thus I
am hesitant to use nbreg, a very different model. Is there any way to get
around this problem under clarify and retain the original model?
Thank you,
Jose Aleman
PhD Candidate
Politics Department
130 Corwin Hall
Princeton, NJ 08544
609.937.0190
609.258.2147
<http://www.princeton.edu/~jaaleman>