Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: -outreg2- for nonlinear regressions?


From   Cindy Gao <cindy.gao@ymail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: -outreg2- for nonlinear regressions?
Date   Fri, 12 Feb 2010 06:55:27 +0000 (GMT)

I am trying to use -outreg2- to export some basic results to excel, for nonlinear regressions. It does not work and I want to know whether outreg2 can be use for nonlinear regressions and if so what I do wrong.

I have basic nonlinear specification like:
nl (GDP= {b0} + {b1}*var1+ var2*({b2} + {b3}*var1) + {b4}*var3+ {b5}*var4)

(don't worry too much about the specification).

Then I say:
outreg2  using newfile ,  stats(coef pval) replace excel

(I only need the coefficients and p-values)

Instead of getting a table of results in excel, I just get something like:

(1)(2)(3)(4)(5)(6)
VARIABLESb0b1b2b3b4b5
Observations3636363636
R-squared0.7970.7970.7970.7970.797
Standard errors in parentheses
*** p<0..01, ** p<0.05, * p<0.1



(but in table format)

It shows no coefficients or p-values, just repeats the # observations and Rsq again and again for each variable. If I regress the same variables using a basic -reg- and then use identical outreg2 command, it works perfectly - making me think maybe outreg2 cannot work for nonlinear regressions? Or what should I do differently?

PS I also try use estout for the same thing, it works fine and exports the results, the problem is that when I append (as I run many regressions that I want to compare) it lists them one below the other but I need them side by side like in outreg2, and I don't know how to store estimates and all that to have them side by side in the table using estout.

I highly appreciate your generous assistance.

Cindy Gao


      


*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index