David Parsley --,
you might want to check the paper "Exports and Productivity Growth –
First Evidence from a Continuous Treatment Approach" by Helmut Fryges
and Joachim Wagner (Working Paper Version available at
http://www.uni-lueneburg.de/fb2/vwl/papers/wp_49_Upload.pdf).
-------------
Abstract: A recent survey of 54 micro-econometric studies reveals that
exporting firms are more productive than non-exporters. On the other
hand, previous empirical studies show that exporting does not
necessarily improve productivity. One possible reason for this result
is that most previous studies are restricted to analysing the
relationship between a firm's export status and the growth of its
labour productivity, using the firms' export status as a binary
treatment variable and comparing the performance of exporting and
non-exporting firms. In this paper, we apply the newly developed
generalised propensity score (GPS) methodology that allows for
continuous treatment, that is, different levels of the firms' export
activities. Using the GPS method and a large panel data set for German
manufacturing firms, we estimate the relationship between a firm's
export-sales ratio and its labour productivity growth rate. We find
that there is a causal effect of firms' export activities on labour
productivity growth. However, exporting improves labour productivity
growth only within a sub-interval of the range of firms' export-sales
ratios.
---------------------
I am not quite sure if their approach identifies the average treatment
effect in the population or just the treatment effect on the treated,
but it may be a starting point. As far as I know they also used Stata
for estimation, so it might be worthwhile to contact them.
Best regards,
Nils
On 7/30/07, David Parsley <[email protected]> wrote:
> Dear Statalist,
>
> I am trying to model the effect of a treatment (x-variable) on an
> outcome (y-variable), where the treatment is a self-selected, CONTINUOUS
> variable (e.g., hours of study, dollars spent, etc.), and the outcome is
> also continuous (performance). I am interested in the effect of the
> treatment on the outcome in the general population - not just the
> treatment sample. I observe y for the entire population (i.e.,
> including those studying/spending, and those not). My treatment is
> either zero, or some positive value (lots of clustering at zero).
>
> I have searched through econometrics texts/articles and (~) 99.99% of
> them focus on self selection of the y-variable. In my case, since the
> selection is on the (continuous) x-variable, including the non-treatment
> population would result in clustering at zero, and therefore biased
> estimates. I have found discussions of treatment effects for dichotomous
> independent variables, but not for 'simple' cases of continuous (though
> strictly positive) treatments.
>
> Treatreg seems to apply only to dichotomous varibles.
>
> Thanks for any help.
>
> David Parsley <http://mba.vanderbilt.edu/david.parsley/>
> Owen Graduate School <http://mba.vanderbilt.edu/>
> Vanderbilt University <http://mba.vanderbilt.edu/>
> Nashville, TN 37203 <http://www.nashville.gov/flashpgs/flashhome.htm>
> (615) 322-0649
>
> *
> * 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/
