Re: st: ETA Stata 12.0

[Neil Shephard <nshephard@gmail.com>]

On Wed, Apr 6, 2011 at 8:17 PM, Charles Koss <hqtiger@gmail.com> wrote:
> Well well! Since starting to use Stata, I am very impressed how much
> have learnt. Since we always want to purchase better products, I hope
> Stata 12 can produce complete inference for Vector Error Correction
> Models (vec command).
>
> Currently, Stata 10 and 11 can produce the estimates for the impulse
> response functions but it CAN NOT estimate the corresponding standard
> errors. So, those estimates are useless since WE DO NEED the standard
> errors for interpreting the results STATISTICALLY. I was amazed to
> learn that OPEN SOURCE software can do that job very efficiently. I
> would like to see this feature in STATA 12.
>
> With this feature, I can convince myself and the administrators that
> the purchase of Stata 12 is worthy, although version 11 is much better
> than 10. If someone has information that contradicts my opinion,
> please do share it.  See:
> http://www.stata.com/statalist/archive/2011-03/msg01435.html

There is nothing stopping you from programming this yourself as an
ado-file if you're that desperate for it.

The -vec- command is implemented as an ado (see -which vec-) which you
could copy to your local ado-path as vec2.ado and hack away at to get
what you want.

If you succeed it would no doubt be useful to submit -vec2- to the SSC archives.

This is likely how the "open source" software that you refer to
(presumably R) has such functionality.

Neil


-- 
“Truth in science can be defined as the working hypothesis best suited
to open the way to the next better one.” - Konrad Lorenz

Email - nshephard@gmail.com
Website - http://kimura.no-ip.org/
Photos - http://www.flickr.com/photos/slackline/

*
*   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/