Re: st: Dynamic almost ideal demand system

[Jorge Eduardo Pérez Pérez <jorge_perez@brown.edu>]

I got confused by the paper by Ray in a previous thread. You provided
the full reference for
 here but it did not look like so. Should have been

Ray (1984) A dynamic generalisation of the almost ideal demand system

In any case, I looked at the papers Ray cites in the first paragraph
and both base inclusion of the lagged values of expenditures on habit
formation equations formulated in levels, so to be strictly consistent
with those papers the lagged values should be included in levels.

Including them in logs may make more econometric sense and provide a
better fit, but it will not be strictly consistent with those papers.
You should think whether the habit formation equations can be
rewritten such that the demand equations include lagged log
consumption, by a change of scale in coefficients perhaps. If so,
there shouldn't be a problem.






--------------------------------------------
Jorge Eduardo Pérez Pérez
Graduate Student
Department of Economics
Brown University


On Fri, Mar 7, 2014 at 4:49 PM, Nigussie Tefera
<nigussie.tefera@gmail.com> wrote:
> Dear stat list
>
> In a dynamic generalization of the the almost ideal demand system,
> Ray(1984) modeled total food expenditure in log terms (lntfexp) whilst
> its lag without log transformation (tfexp_1); also lag of expenditure
> for each food groups is without long transformation. For practical
> estimation, I want to make sure whether the log or alternative
> (absolute terms) should be considered or not. Your suggestion is very
> welcome.
>
>
> Best
>
> Nigusise
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/