RDSURVIV fails to execute with large data set

questions concerning analysis/theory using software from the Patuxent Software Archive

RDSURVIV fails to execute with large data set

Postby Todd » Thu Jan 29, 2009 4:53 pm

We are getting the following error when trying to run RDSURVIV to asses GoF for a Robust Design model that was constructed and run in Program MARK.

RDSURVIV wrote:Number of classes exceeds MXCLAS. Recompile with MXCLAS=# of classes.

Unfortunately, we can't find anything in the RDSURVIV manual that describes this error, the MXCLAS statement, or how to "recompile" the data.

We have a MARK encounter history file with each animal being represented by a single line and we used CNVRDSURVIV to convert the encounter history file into summary mij arrays.

Can anyone direct us to a solution?
Todd
 
Posts: 20
Joined: Thu Feb 10, 2005 2:07 pm

RDSURVIV problem

Postby ganghis » Thu Jan 29, 2009 5:08 pm

You've exceeded the maximum dimensions that RDSURVIV allows for 1 or more of it's internal vectors- you'll probably need Jim Hines to recompile it for you before it'll work. To make his life easier, it would be good to post the dimensions that you need (i.e., #occasions, groups, etc.).

Cheers, Paul
ganghis
 
Posts: 84
Joined: Tue Aug 10, 2004 2:05 pm

Postby Todd » Thu Jan 29, 2009 5:17 pm

In our original MARK .inp file, we had 2 groups (male or female) and 8 primary sampling occasions with secondary samples of 10-10-10-10-10-6-6-6 for a total of 68 encounter possibilities. We had 411 total animals.

It did not appear from the manual that RDSURVIV could handle more than one group so we simply dispensed with the groups and coded every animal as either alive '1' or removed/dead '-1'.

We used CNVRDSURVIV to create the mij arrays.
Todd
 
Posts: 20
Joined: Thu Feb 10, 2005 2:07 pm


Return to analysis help

Who is online

Users browsing this forum: No registered users and 9 guests

cron