Multistate model bug?

questions concerning analysis/theory using program MARK

Multistate model bug?

Postby Morten Frederiksen » Wed Aug 08, 2007 8:45 am

I'm trying to change which parameter to estimate by subtraction in a multistate model. This facility is very helpful, particularly as it allows you to build useful additive models. However, in my version of MARK (4.3) it doesn't quite seem to work. I can change the probability to obtain by subtraction in the dialog box, but MARK seems to take no notice of this when I run the model. In fact, if I immediately reopen the dialog box, the default values are shown rather than what I just chose. What's going on - is it a bug?

Morten
Morten Frederiksen
 
Posts: 42
Joined: Thu May 29, 2003 4:53 am
Location: Roskilde, Denmark

Re: Multistate model bug?

Postby cooch » Wed Aug 08, 2007 9:19 am

Morten Frederiksen wrote:I'm trying to change which parameter to estimate by subtraction in a multistate model. This facility is very helpful, particularly as it allows you to build useful additive models. However, in my version of MARK (4.3) it doesn't quite seem to work. I can change the probability to obtain by subtraction in the dialog box, but MARK seems to take no notice of this when I run the model. In fact, if I immediately reopen the dialog box, the default values are shown rather than what I just chose. What's going on - is it a bug?

Morten


Can't replicate the problem. The change shows up in terms of which parameters are presented in the real and beta outputs. You don't really see anything otherwise.
cooch
 
Posts: 1654
Joined: Thu May 15, 2003 4:11 pm
Location: Cornell University

Multistate model bug?

Postby gwhite » Wed Aug 08, 2007 11:29 am

Morton:
Check the PIM Chart labels, or else the PIM Window labels, and I think you will see that the change was made. The default values in the change window are not changed.

Gary
gwhite
 
Posts: 340
Joined: Fri May 16, 2003 9:05 am

Postby Morten Frederiksen » Mon Aug 13, 2007 4:01 am

Thanks Gary and Evan - it does indeed work, although it's very discreet!

Morten
Morten Frederiksen
 
Posts: 42
Joined: Thu May 29, 2003 4:53 am
Location: Roskilde, Denmark


Return to analysis help

Who is online

Users browsing this forum: No registered users and 5 guests