Wir sind überzeugt, dass Sie unsern SQLArtist testen sollten!
Mit dem SQLArtist verwalten Sie auch komplexe Datenbanken mühelos und sicher.
- Editieren von DB-Strukturen
- SQL-Scripts erstellen (Wizard)
- DB-Versions-Verwaltung
- Möglichkeit der Anbindung an Borland Delphi und PowerDesigner
- und noch viele weitere Funktionen
URL zu Produktebeschreibung und Download Vollversion:
http://www.vid-software.com/prod_sqlartist.php
Mit freundlichen Grüssen
VID-Software GmbH
Stefano Marinello
Hofstrasse 99
8620 Wetzikon
Switzerland
-----------------------------------------------------------------------
Um sich von dieser Mailingliste abzumelden
http://www.vid-software.com/newsletter/lists/?p=unsubscribe&hash=1db757e191…
Um Ihre Benutzerdaten zu manipulieren
http://www.vid-software.com/newsletter/lists/?p=preferences&uid=1db757e1914…
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.hmdc.harvard.edu/?info=clarify
Karl,
weibulls can be difficult to interpret because sometimes huge numbers get
(randomly) drawn from the distribution. these outlier draws can result in
means that are much much larger than anticipated. there are two ways you
can check your work. first, try looking at the median rather than the mean.
second, to see whether clarify is at fault, try using stata's "predict"
command to generate predicted values. this command does not produce the
same output as clarify (at least not in its default mode), but if you
continue to get nonsensical results at least you will have narrowed the
source of the problem.
best wishes on your research!
Jason
At 08:27 AM 2/4/2003 +1300, you wrote:
>jason: thanks for getting back to me. i eventually realized what the
>problem was: clarify does not support 'streg' stata command but it does
>support the 'weibull' command (i didnt know there was a weibull command).
>i tried estsimp with competing risks weibull and it worked but i got
>really high predicted duration times that made no sense. maybe my models
>are bad. ...
>
>so now i am trying to figure out how to do competing risks followed by
>clarify. for some reason i think problem may be if there are only a few
>failures (in my case as low as 16 in a sample of 90) it becomes trickier.
>any thoughts?
>
>regards,
>
> -----Original Message-----
> From: Jason Wittenberg [mailto:jwittenberg@wcfia.harvard.edu]
> Sent: Tue 4/02/2003 8:16 a.m.
> To: Karl DeRouen
> Cc:
> Subject: Fwd: follow-up
>
>
>
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.hmdc.harvard.edu/?info=clarify