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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: AW: confidence interval of a ratio of coefficients


From   Arne Risa Hole <arnehole@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: AW: confidence interval of a ratio of coefficients
Date   Fri, 9 Apr 2010 17:52:56 +0100

Rodolphe's suggestion calculates a parametric bootstrap CI, so this is
essentially the same as using the -krinsky- option of -wtp-. The two
approaches will not give *exactly* the same results, however, as the
draws from the multivariate normal distribution are not generated in
the same way.

Arne

On 9 April 2010 16:34, Rodolphe Desbordes
<rodolphe.desbordes@strath.ac.uk> wrote:
> Oops... the "qui" are of course unnecessary...
>
> Rodolphe
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Rodolphe Desbordes
> Sent: vendredi 9 avril 2010 16:26
> To: 'statalist@hsphsun2.harvard.edu'
> Subject: RE: st: AW: confidence interval of a ratio of coefficients
>
> Martin and Nick,
>
> Very true:
>
> ***
> sysuse auto, clear
> set seed 2940688
> logit foreign price weight rep78
>
> preserve
>
> drawnorm b1-b4, n(20000) means(e(b)) cov(e(V)) clear
>
>
> gen ratio=-(b2)/(b1)
>
> qui sum ratio,meanonly
>
> dis "ratio: " r(mean)
>
> qui _pctile ratio, p(2.5,97.5)
>
> dis "ll: " r(r1)
> dis "ul: " r(r2)
>
>
>
> restore
>
> ***
>
> Rodolphe
*
*   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