<>
" It uses the delta-method estimate for the variance of
exp(b)."
The variance/sd did, however, coincide with the one returned by -xtmelogit-.
Thanks for the advice on the endpoints anyway...
HTH
Martin
-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von
[email protected]
Gesendet: Donnerstag, 12. November 2009 03:19
An: [email protected]
Betreff: Re: st: Intraclass correlation and binary hierarchical models
Martin, -nlcom- doesn't invert the endpoints of the log-scale
interval. It uses the delta-method estimate for the variance of
exp(b). It's not a bug, but a trade-off for the great generality of
expressions that -nlcom- can take.
-Steve
On Wed, Nov 11, 2009 at 4:51 PM, Martin Weiss <[email protected]> wrote:
> The ci returned by -nlcom- is off slightly, I do not know why, though.
>
> nlcom (sd: exp(_b[lns1_1_1:_cons]))
>
*
* 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/
*
* 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/
