st: nl command / no intercept

["Christoph Neumann" <c.neumann@tu-clausthal.de>]

Dear Statalisters,

I want to do a nonlinear regression without an intercept in the model. I was
surprised about the high R² values in my solution and had a look into the
literature and found the discussion about the R² <=> model without intercept
problem. For a linear regression there is the suggestion to use the
non-modified R² and the option tsscons (And use the R² values with a
different interpretation). For the nonlinear case I couldn´t find any
advices. Furthermore there is no option tsscons within the nl command.

Any suggestions or experiences? Thanks a lot!

Best,
Christoph
Technical University Clausthal (using Stata 12, Windows),
c.neumann@tu-clausthal.de


*
*   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/