Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.

# AW: AW: st: prgen option quadratic term

 From "Meulemann Max" To "statalist@hsphsun2.harvard.edu" Subject AW: AW: st: prgen option quadratic term Date Wed, 7 Nov 2012 13:47:27 +0000

```I also did the regressions with centered variables, but the picture essentially stays the same.

y is an assement of importance on a certain burden sharing rule, values are not important, mod. imp, important, very imp.

I am also getting the feeling that the result is driven by outliers in capitagdp. Thx for the help, I should be able to interpret it myself, just seemed wired for me personally that there would be a quadratic term and a turning point on that issue im looking at.
________________________________________
Von: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu]&quot; im Auftrag von &quot;Daniel Feenberg [feenberg@nber.org]
Gesendet: Mittwoch, 7. November 2012 12:39
An: statalist@hsphsun2.harvard.edu
Betreff: Re: AW: st: prgen option quadratic term

> On Wed, Nov 7, 2012 at 10:35 AM, Meulemann  Max <mmeulemann@ethz.ch> wrote:
>> thx, the margins and marginsplot function helped a lot.
>>
>> Anyways anybody got a idea how I am to interpret my data here: I find a
>> negativ linear term for capitagdp, but if i include a quadratic gdp
>> term, that one turns out to be positiv and significant with a lrtest.

If you rerun your regression, but replace gpd*^2 with the square of gdp
less mean gdp, then you have basically the same model and standard errors,
but the coefficient on the squared term will have no influence on the
dependent variable at the mean gdp. So you have an easy interpretation of
the linear term - it is the effect at the mean gdp. The sign of the
quadratic coeficient then tells you if the effect grows or shrinks for gdp
larger or smaller than the mean and it is easy to see how intense that
effect will be. The other coeficients will be unaffected.

dan feenberg
NBER
*
*   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/
```