At 16.02.2009 16:47, you wrote:
--- On Mon, 16/2/09, Michael Stobernack wrote:
> Another possibility could be to incorporate dummy variables
> and interaction terms for the different groups of interest.
> Since I have got four different groups I need three dummies
> and many interactions terms, so I loose (to) many degrees of
> freedom (n=50).
By estimating separate models for each group you actually
estimate 4 extra coefficients, i.e. loose 4 degrees of
freedom,
I estimate the following model 4 times, i.e. once for each of the four groups:
y = b0 + b1*x1 + b2 * x2 + b3*x3 + e, so I have to estimate 4*4 = 16
coefficients.
If I estimate one model for all 4 groups together and D1 is a dummy
four group1 etc, I use
y = b0 + b1*x1 + b2*x2 + b3*x3 +
b4*D1 + b5*D2 + b6*D3 + (to
see, whether the intercept differs between the groups)
b7*D1*x1 + b8*D2*x1 + b9*D3*x1 + (to see,
whether the slope of x1 differs between the groups)
b10*D1*x2 + b11*D2*x2 + b12*D3*x2 + (to see,
whether the slope of x2 differs between the groups)
b13*D1*x3 + b14*D2*x3 + b15*D3*x3 + epsilon (to see,
whether the slope of x3 differs between the groups)
So I have to estimate 16 coefficients.
What do you mean by 'estimate 4 extra coefficients'? Do you mean 4
extra values of the residual variance?
Michael
compared to a model with all the interaction
effects. The reason is that in the interaction model you
constrain the residual variance to be equal across groups.
-- Maarten
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room N515
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* 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/
