Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Fwd: FW: output tables after xtlogit using outreg2


From   Abhimanyu Arora <abhimanyu.arora1987@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Fwd: FW: output tables after xtlogit using outreg2
Date   Mon, 4 Oct 2010 06:52:08 +0200

Dear statalist
With regard to my post
http://www.stata.com/statalist/archive/2010-10/msg00012.html, please
find below the solution.
Best regards
Abhimanyu


---------- Forwarded message ----------
From: Abhimanyu Arora <abhimanyu.arora@student.kuleuven.be>
Date: Mon, Oct 4, 2010 at 6:45 AM
Subject: FW: output tables after xtlogit using outreg2
To: "abhimanyu.arora1987@gmail.com" <abhimanyu.arora1987@gmail.com>




________________________________
From: Roy Wada [roywada@hotmail.com]
Sent: Monday, October 04, 2010 6:23 AM
To: Abhimanyu Arora
Subject: RE: output tables after xtlogit using outreg2

Hi Abhimanyu,

Below is the answer to your question. Do me a favor and send it to the
Statalist. It will save me from having to answer the same question in
the future.


You want to match variables across equations. There a few ways to do
this. There is an undocumented option called -eq(auto)-. This works
95% of time.

cd c:/
use union, clear
xtset idcode year
xtlogit union age grade i.not_smsa south##c.year,
outreg2 using table,ctitle(RE) word replace eq(auto)
xtlogit union age grade i.not_smsa south##c.year, fe
outreg2 using table,ctitle(FE) word append eq(auto)


Another way is to create a sub-option called -eqmatch-. You can just
stick it in to the top of your do-file. In the following example, the
equation rep78 is matched to the equation trunk.

prog drop _all
prog define eqmatch
 syntax anything
 gettoken one two: anything
 local one=trim("`one'")
 local two=trim("`two'")
 split varname, parse(":") gen(_varname)
 replace eqname="`one'" if eqname=="`two'"
 replace report=eqname+":"+_varname2
end

sysuse auto, clear
ologit rep78 foreign mpg
outreg2 using myfile, replace nocons
ologit trunk foreign mpg
outreg2 using myfile, ststr(eqmatch rep78 trunk) nocons


The third way is use -eqmatch(   )- option. This is not avail on the
public version because I didn't feel like updating it ssc.


Thanks,

Roy



> From: abhimanyu.arora@student.kuleuven.be
> To: roywada@hotmail.com
> Subject: output tables after xtlogit using outreg2
> Date: Sat, 2 Oct 2010 16:43:10 +0200
>
> Dear sir
>
> I am Abhimanyu Arora, PhD student in economics at the KU, Leuven, Belgium. I
> shall be grateful if you could offer some guidance regarding the use of
> outreg2 command after xtlogit commands (once for random effect and once for
> fixed effect).
>
> I type the following commands
>
> xtlogit depvar indvars
> outreg2 using table,ctitle(RE) word replace
> xtlogit depvar indvars,fe
> outreg2 using table,ctitle(FE) word append
>
> Since the predictors are the same in both cases, ideally I would like to
> have one table with two columns with the rows corresponding to each
> predictor. However, I get a 2-columned table but the predictors are repeated
> twice with one half of the now doubly long table empty for each column. The
> problem did not exist for other estimation commands (reg/qreg) for auto.dta.
>
> Is it the case that oureg2 is for certain estimation commands only?
>
> Please find attached the output word file. I do not understand what does
> "EQUATION SINGLE" means in the first column. You might also notice that the
> column for coefficients for random effect is not named RE.
>
> I sent the query to statalist yesterday too, but wasn't able to get any
> response
>
> Thanking you in anticipation
> Yours sincerely
> Abhimanyu Arora
>
>
> LICOS - Centre for Institutions and Economic Performance
> & Research Group on Development Economics
> Faculty of Business and Economics, K.U.Leuven
> Deberiotstraat 34 - bus 3511
> B-3000 Leuven
> http://www.econ.kuleuven.be/licos
> Tel 00 32 (0) 16 32 65 66
> Fax 00 32 (0) 16 32 65 99
>
>

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index