Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: 2 stage regression


From   [email protected]
To   [email protected], Mindell Nitkin <[email protected]>
Subject   st: Re: 2 stage regression
Date   Thu, 17 Nov 2005 08:34:14 -0500

Quoting Mindell Nitkin <[email protected]>:

Thanks for the suggestion, but that isn't quite what I am looking for, so I'll
try to clarify.

I want to do industry based regressions in year t-1, then use the coefficients
from the industry regressions to predict the value of accrual for the firms in
each industry in year t. I am using 2 digit SIC codes to define the industries.
I have roughly 50 industry groupings with anywhere between 10 and 100 firms per
industry.

For example,
In the first stage:
         by SIC: regress accruals PPE REV AR INV
In the second stage, I want to use the coefficients of PPE REV AR and INV to
predict accruals and to determine the unexpected portion of accruals (the
residuals) for each firm within each industry.

Again thanks for the help,
Mindy Nitkin
DBA candidate
Boston University



 Michael Blasnik wrote:
>
> > "Caleb Southworth" <[email protected]>  wrote:
> >
> >> Within SIC lags:
> >> sort SIC
> >> by SIC: gen CF_1=CF[_n-1]
> >> by SIC: gen CF_2=CF[_n-2]
> >> etc.
> >>
> >
> > This approach should not be used to create lagged values because Stata
> > randomly breaks ties during a sort.  You must specify the time
> > variable using -bysort SIC (time): gen ...- or else use Stata's
> > built-in time series operators which are much smarter and more careful.
> >
> > Michael Blasnik
> > [email protected]
> >
> > *
> > *   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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index