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.
