--- Heckendorn Felix <[email protected]> wrote:
> The problem in fact might appear simple. I have an experiment with 3
> different treatments (one of which is the control treatment). The
> response variable follows a negative binomial distribution.
> The experiment was conducted at 6 different sites which gives us 6
> replicates (n=6).
> What I ultimately want to know is:
> 1. how much better is my treatment 1 and 2 when compared to the
> control.
> 2. is there a significant influence of the experimental sites.
>
> I recoded my treatments and have set the control treatment to 0. My
> model then consists of my response and two explanatories (i)
> treatment and (ii) experimental site. With the gnbreg command I get
> coefficients for the 2 treatments and for 5 experimental sites (one
> site omitted). If I want to know whether experimental site as such is
> an important variable for the model, I need to check the model with
> and without the variable 'experimental sit'. Apparently the residual
> deviance D allows an answer to that question. How to do?
This is implemented as the likelihood ratio test, see -help lrtest-.
Alternatively you can do the Wald test, see -help test-.
-- 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 Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
__________________________________________________________
Not happy with your email address?.
Get the one you really want - millions of new email addresses available now at Yahoo! http://uk.docs.yahoo.com/ymail/new.html
*
* 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/
