Re: st: interpreting coefficients in growth equation

["D-Ta-Fresh" <altruist81@gmx.de>]

Yes, excactly. And in fact im interested in heterogenous effects of 
income on BMI growth looking at different sub-samples in the data.

Still, im not sure if thats the best way to model it as I have a hard 
time to interpret the logIncome coefficient ( I know what it is, but 
find it hard to get an idea of what it actually means)

Darjusch

-------- Original-Nachricht --------
> Datum: Sun, 27 Mar 2011 06:33:37 -0500
> Von: Charles Koss <hqtiger@gmail.com>
> An: statalist@hsphsun2.harvard.edu
> Betreff: Re: st: interpreting coefficients in growth equation

> Sorry, my eyes did not see the negative sign. In this way, on average,
> a unit change in higher initial values of BMI leads to lower levels of
> growth. Now, i am intrigued. Basically, you are regressing the growth
> rate by  the previous level (you called it present), logincome, and
> age.  is it logincome and age measured at the previous level (you
> called it present) or current period ?
> 
> Charles
> 
> -- 
> Charles Koss
> http://charlesonnet.blogspot.com
> 
> 
> On Sun, Mar 27, 2011 at 4:44 AM, D-Ta <altruist81@gmx.de> wrote:
> > b1 = -0.33, i.e. A higher initial level of BMI should lead to lower
> growth
> > rates, isnt it?
> >
> >
> > Am 26.03.2011 20:07, schrieb Charles Koss:
> >>
> >> on average, an additional year changes BMI_growth by 0.11, b2 might be
> >> an elasticity  (1% change in logincome causes 0.14% in BMI_growth ).
> >> b1,  a unit change from the average present level of BMI  changes the
> >> BMI_growth by 0.33; in other words, increases the growth of BMI. So, i
> >> think you may have divergence rather than convergence.
> >>
> >> Does it make sense? otherwise, provide more details about the dataset.
> >>
> > *
> > *   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/
> >
> 
> *
> *   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/

-- 
GMX DSL Doppel-Flat ab 19,99 Euro/mtl.! Jetzt mit 
gratis Handy-Flat! http://portal.gmx.net/de/go/dsl
*
*   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/