Thanks to Ben Jann and Kit Baum! It really helped and the changed format
did fine! I agree with Kit Baum but for other reasons I had to use estout
- if I use internally saved results is there a vector with variable names
saved as strings?
Thanks,
Johannes
This is not an estout problem. Files produced by estout or estimates
table are not meant for further computation; they're created for
presentation. If you want to work with the coefficient vectors and
standard errors of these estimates, save the matrices:
logit y x1 x2 x3
mat b1 = e(b)
mat V1 = e(V)
logit y2 x4 x5 x6
mat b2 = e(b)
mat V2 = e(V)
You may then do whatever numerical calculations you wish with the
results at the full numerical precision with which they are stored.
Although displaying a matrix (e.g. mat list b1) will use a default
format, any calculations will use full accuracy.
You may also access these matrices' contents from within Mata if you
want to carry out further calculations in a powerful matrix language.
Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html
Try
... cells("b(fmt(%18.8f)) se( fmt(%18.8f))") ...
(Although I think it makes more sense to use the g-format here, say
%12.0g.)
ben
On Apr 24, 2007, at 2:33 AM, statalist-digest wrote:
>
> I have a problem with the command "estout" - I would like to use
> regression results for further calculations, I need variable names,
> coefficients and standard errors with at least 8 decimal places,
> e.g., the
> following code:
>
> logit y1 x1 x2 x3
> estimates store reg1
>
> logit y2 x4 x5 x6
> estimates store reg2
*
* 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/
*
* 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/
