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 at the end of May, and its replacement, statalist.org is already up and running.


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

st: orders of operation for a seasonal difference quadratic regression using factor variables


From   "Dimitriy V. Masterov" <dvmaster@gmail.com>
To   Statalist <statalist@hsphsun2.harvard.edu>
Subject   st: orders of operation for a seasonal difference quadratic regression using factor variables
Date   Wed, 8 Aug 2012 16:25:04 -0700

I have monthly panel data on unemployment (uerate) and an outcome of
interest (lny) by US county. I ran the following 12-month seasonal
difference regression with a quadratic unemployment term (lagged one
month to give my counties time to adjust):

regress S12.(lny cL1.uerate##cL1.uerate) i.countyid#c.dateym, vce(cluster state)

My (misguided) intuition is that Stata would expand the factor
variables and then do the seasonal differencing, resulting the OLS
regression:

lny_{t}-lny_{t-12} = a + b*(x_{t-1}-x_{t-13}) + c *
(x_{t-1}^2-x_{t-13}^2) + county-specific trend.

Stata actually seems to be estimating:

lny_{t}-lny_{t-12} = a + b*(x_{t-1}-x_{t-13}) + c *
(x_{t-1}-x_{t-13})^2 + county-specific trend.

The manual says that D. and S. operators may not be combined with
factor variables. It provides a categorical variable example, which
makes sense, but does not apply here. There's no error message and
Stata actually estimates something, but I can't seem to find the logic
anywhere in the pdf documentation.

DVM
*
*   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