But in roa1 i don`t need all the observations
example for a firms with information of roa for 1990-2004 I only will have
roa1 for 1990-2002 and for a firm with roa for 1990-1994 I`ll have have info
for 1990-1992
-----Original Message-----
From: "Eric G. Wruck" <ewruck@econalytics.com>
To: statalist@hsphsun2.harvard.edu
Date: Tue, 27 Jun 2006 10:42:16 -0400
Subject: Re: st: superposed rolled windows
> That's why Jeph had the statement
>
> drop if profit2 == .
>
> except that I think it should have read
>
> drop if profits3 == .
>
>
> This will include only those observations that have a profit for this
> year & the following two. Per Nick's recommendation, you should also
> give some serious thought to gaps in profit data for a given firm.
>
> Eric
>
>
> p.s. Jeph, what is the meaning of HTH?
>
>
> >But i dont want all the observations
> >for example if a firm has observations for 1990-2004 the new database
> will
> >have observations for 1990-2002. In the same way for a firm with data
> for
> >1990-1995 will have observations for 1990-1993
> >Thanks
> >
> >-----Original Message-----
> >From: Jeph Herrin <junk@spandrel.net>
> >To: statalist@hsphsun2.harvard.edu
> >Date: Tue, 27 Jun 2006 10:17:23 -0400
> >Subject: Re: st: superposed rolled windows
> >
> >> Try
> >>
> >> gen profits1=profits
> >> bys firms (years): gen profits2=profits[_n+1]
> >> bys firms (years): gen profits3=profits[_n+2]
> >> drop if profit2==.
> >>
> >> HTH,
> > > Jeph
> >>
>
>
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/