# Re: st: Post again: questions on difference models

 From Jeph Herrin To statalist@hsphsun2.harvard.edu Subject Re: st: Post again: questions on difference models Date Fri, 14 Jul 2006 17:03:54 -0400

You may want to post this to a statistics list, this is a list
concerning the use of Stata.

cheers,
Jeph

Carol Bao wrote:
Hi, there,

I am running a first differencing model and have two
estimation questions:

1. For a dichotomous dummy variable, the difference
can be -1, 0 and 1. Instead of using this directly as
a regressor, is it econometrically correct to generate
three new dummy variables respectively for the three
categorized difference and use them as regressors

eg. instead of: \delta Y = \beta \delta X,
estimate: \delta Y = \beta1 Z2 + \beta2 Z3, where Z2=1
if \delta X==0 and Z3=1 if \delta X==1 (default is if
\delta X=-1).

If yes, is there any paper or book chapter that can be
referred to?

2. Dealing with age and year dummies. Since the
difference age is the same for all persons, it seems
meaningless to use (\delta age) as a regress (or does
it?) Is it correct to use individual age dummies to
control for cohort effects? Also, should year dummies
be included as well? Is there any paper or book
chapter that can be referred to in this regard?

Any information is highly appreciated.
Thank you very much in advance,

Carol

__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com *
* 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/