Laszlo Sandor <[email protected]> is using -moptimize()- and found that the
documentation is referring to -moptimize_result_coefs()- instead of
-moptimize_result_eq_coefs()-:
> the help (e.g. http://www.stata.com/help.cgi?mf_moptimize#syn_results)
> says to access the results with the function -
> moptimize_result_coefs(M [, i]) -. This did not work for me. From the
> initialization nomenclature, I got inspiration, and tried
> -moptimize_result_eq_coefs(M [, i]) -. This works.
>
> It could have saved days of my life.
Laszlo is correct. The documentation fails to mention the
-moptimize_result_eq_coefs()- function, and refers to
-moptimize_result_coefs()- in its place. The documentation should read
something like the following:
moptimize_result_eq_coefs(M [, i]) returns the 1 x (ki+ci) coefficient
rowvector for the ith equation. If i>=. or argument i is omitted, the
1 x K full set of coefficients is returned.
moptimize_result_coefs(M) returns the 1 x K full set of coefficients.
We will fix this in a future ado-file update.
> By the way, I dropped a note here months ago that the very same help
> file is inconsistent in notation about initializing values for the
> coefficient vector, whether or not the constant term is meant to be
> initialized by -moptimize_init_eq_coefs(M, i, b0)- or not. It should.
> The helpfile has not been updated accordingly.
Although not explicitly mentioned in -help whatsnew-, the help file for
-moptimize()- was updated regarding -moptimize_init_eq_coefs()- on 08dec2009.
--Jeff
[email protected]
*
* 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/
