Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Identifying changes in the effect of a variable over time?


From   Kam Kup <[email protected]>
To   [email protected]
Subject   st: Identifying changes in the effect of a variable over time?
Date   Wed, 7 Feb 2007 07:57:33 -0800 (PST)

I have panel data on choice.  My panel is fairly
short, 4 observations on each of the independent and
dependent variables. My data, of course, is in long
form with "t" as the time variable.

I'd like to identify changes in the effect of certain
independent variables over time.  The way I did this
was by establishing dummy variables for three out of
four of the time periods.  For examples

gen dummytime2=0
replace dummytime2=1 if t=2

To try to identify changes in the effect of an
independent variable over time, I've interacted the IV
with these round dummies.  For example, income is a
variable that does not vary over time in this study. 
So I've established an interaction:

gen Incometime2=dummytime2*income
gen Incometime3=dummytime3*income
gen Incometime4=dummytime4*income

and the command would be:

xtprobit Choice Income Incometime2 Incometime3
Incometime4 [other variables] dummytime2 dummytime3
dummytime4

My hope is to identify changes in the effect of income
in each time period.  The last three covariates are
meant to identify time-specific constants.

Is this correct? If not, is there a better way of
identifying these changes over time?

Thanks,
Kam


 
____________________________________________________________________________________
It's here! Your new message!  
Get new email alerts with the free Yahoo! Toolbar.
http://tools.search.yahoo.com/toolbar/features/mail/
*
*   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