Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: rename estimate coefficients: ereturn repost
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: rename estimate coefficients: ereturn repost
Date
Wed, 23 Jan 2013 18:56:31 +0000
If I understand what you are trying, it works for me. Here is a program
program mydisp, eclass
matrix betass=e(b)
matrix colnames betass = "aa" "bb" "cc" "DD" "ee" "ff"
mat list betass
ereturn repost b = betass, rename
end
and this uses it
sysuse auto, clear
gen gpm = 1/mpg
reg gpm displacement headroom weight turn trunk
mydisp
mat b = e(b)
mat li b
and this is what I get
. mat li b
b[1,6]
aa bb cc DD ee ff
y1 4.806e-06 -.00065331 .00001154 .00019261 .00038585 .00339977
Seems simpler just to use the names you want as variable names,
although the constant will still be called _cons.
That said, I can't advise on -outreg2- (SSC), -outreg- (SJ, SSC) and
-estout- (SJ, SSC) except to make a factual comment that the comments
in the -outreg2- documentation on -outreg- don't make sense except in
reference to a now superseded version of -outreg-.
Nick
P.S. On pleas for urgency, see
http://www.stata.com/support/faqs/resources/statalist-faq/#others
On Wed, Jan 23, 2013 at 6:04 PM, Johannes Breckenfelder
<[email protected]> wrote:
> Thank you very much, Aljar, for the comments below. Unfortunately it doesn't solve my problem, it would be awesome if you could have another look:
>
> I'm starting off with these stored estimates:
> ****************************************************************************
> . eret list
>
> scalars:
> e(N) = 534
>
> macros:
> e(cmd) : "regress"
> e(depvar) : "Excess Return"
> e(properties) : "b V"
>
> matrices:
> e(b) : 1 x 5
> e(V) : 5 x 5
>
> functions:
> e(sample)
> ***************************************************************************
>
> Next, I'm running this procedure to change column names:
> ****************************************************************************
> program drop disp2
> program disp2, eclass
> matrix betass=e(b)
> matrix colnames betass = "aa" "bb" "cc" "DD" "ee" "ff"
> mat list betass
> ereturn repost b = betass, rename
> end
> ****************************************************************************
>
> Then, I check whether it actually did change -and it didn't....
> ****************************************************************************
>
> matrix betass=e(b)
> mat list betass
>
> betass[1,6]
> r1 r2 r3 r4 r5 r6
> y1 1.0086253 .76111561 .00520306 .26619872 .02242498 0
> ****************************************************************************
>
> ... even though the individual step does work:
> ****************************************************************************
>
> matrix colnames betass = "aa" "bb" "cc" "DD" "ee" "ff"
> mat list betass
>
> betass[1,6]
> aa bb cc DD ee ff
> y1 1.0086253 .76111561 .00520306 .26619872 .02242498 0
> ****************************************************************************
>
> Somehow it doesn't go through... Any ideas?
Aljar Meesters
> I do not know what you exactly want to do, but in order to get your program running you should:
> 1) rename disp
> disp is already a Stata build-in command.
>
> 2) Not use ereturn post b V
> You do not have a matrix b so you can not ereturn it.
>
> This should do what I think that you have in mind:
>
> program disp2, eclass
> matrix betass=e(b)
> matrix colnames betass = "aa" "bb" "cc" "DD" "ee" "ff"
> mat list betass
> ereturn repost b= betass, rename
> end
2013/1/23 Johannes Breckenfelder <[email protected]>:
>> I would like to create a multi-column table with outreg2 from stored
>> estimates. There are no labels stored with the estimates, just the b
>> V. For each stored set of estimates, I run:
>>
>>
>>
>> program disp, eclass
>>
>> ereturn post b V
>>
>> matrix betass=e(b)
>>
>> matrix colnames betass = "aa" "bb" "cc" "DD"
>> "ee" "ff"
>>
>> mat list betass
>>
>> ereturn repost b = betass, rename
>>
>> end
>>
>>
>>
>> Every step seems to produce the desired outcome, except the last one:
>> ereturn repost b = betass, rename.
>>
>>
>>
>> Where is my coding/conceptual error?
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/