Hi,
I have two questions. I am trying to get a clarify graph showing how the
effective number of parties changes with magnitude depending on how old a
democracy is. I use the following code to generate the graph and get an error
message stated below. Does anyone have a suggestion what's wrong?
generate plo = .
generate phi = .
generate magnitudeaxis = _n + 1 in 1/150
setx age 10 avemag mean elf mean
local a = 1
while `a' <= 150 {
setx avemag `a'
simqi, ev genev(myvariable)
_pctile myvariable, p(.5,99.5)
replace plo = r(r1) if magnitudeaxis==`a'
replace phi = r(r2) if magnitudeaxis==`a'
drop myvariable
local a = `a' + 1
}
sort magnitudeaxis
graph plo phi magnitudeaxis, s(ii) c(||)
Error message: graph plo phi magnitudeaxis, s(ii) c(||)
plograph_g.new phi magnitudeaxis, s(ii) c(||): class member function not found
r(4023);
My second question is related. The model I have chosen to use is a pooled cross-
sectional time-series, which is not supported by clarify. I am not sure how to
set it up in order to simulate all the data separately and then generate the
graphs. Any help would be appreciated.
Thank you,
Ekaterina
--
Ekaterina R. Rashkova
Graduate Student
Department of Political Science
Washington University
Campus Box 1063
One Brookings Drive
St. Louis, MO 63130
(314) 935-5856 (Fax)
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.hmdc.harvard.edu/?info=clarify
Hi Ekaterina,
I think the code you're using for the graph command at the bottom of your file is from an older version of Stata. If you're using Stata 8, try looking into the use of twoway. In your case, I think you would want to type:
twoway rcap plo phi magnitudeaxis
Re. your second question, actually, I think Clarify does support pooled models. What it doesn't support is fixed and random effects models. Then you'd have to do some coding on your own.
Best regards,
Colin
Colin Vance, Ph.D.
German Aerospace Center
Institute of Transport Research
Rutherfordstrasse 2
12489 Berlin
Germany
tel: +49 30 67055147
fax: +49 30 67055202
email: colin.vance(a)dlr.de
-----Ursprüngliche Nachricht-----
Von: owner-clarify(a)latte.harvard.edu [mailto:owner-clarify@latte.harvard.edu] Im Auftrag von errashko(a)artsci.wustl.edu
Gesendet: Tuesday, September 13, 2005 8:59 AM
An: clarify(a)latte.harvard.edu
Betreff: [clarify] question on clarify graphs code
Hi,
I have two questions. I am trying to get a clarify graph showing how the effective number of parties changes with magnitude depending on how old a democracy is. I use the following code to generate the graph and get an error message stated below. Does anyone have a suggestion what's wrong?
generate plo = .
generate phi = .
generate magnitudeaxis = _n + 1 in 1/150 setx age 10 avemag mean elf mean local a = 1 while `a' <= 150 {
setx avemag `a'
simqi, ev genev(myvariable)
_pctile myvariable, p(.5,99.5)
replace plo = r(r1) if magnitudeaxis==`a'
replace phi = r(r2) if magnitudeaxis==`a'
drop myvariable
local a = `a' + 1
}
sort magnitudeaxis
graph plo phi magnitudeaxis, s(ii) c(||)
Error message: graph plo phi magnitudeaxis, s(ii) c(||) plograph_g.new phi magnitudeaxis, s(ii) c(||): class member function not found r(4023);
My second question is related. The model I have chosen to use is a pooled cross- sectional time-series, which is not supported by clarify. I am not sure how to set it up in order to simulate all the data separately and then generate the graphs. Any help would be appreciated.
Thank you,
Ekaterina
--
Ekaterina R. Rashkova
Graduate Student
Department of Political Science
Washington University
Campus Box 1063
One Brookings Drive
St. Louis, MO 63130
(314) 935-5856 (Fax)
--
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