I recently ran known fates models for 5 years of ruffed grouse survival data with sex, age, area, year, and season as grouping variables. Everything looks good, but the model weight for the best model (Year*Season) was 0.9999. Seems hard to believe that this one could be SO much better than all competing models. Anyone had similar results with high model weights? Should the results be viewed as suspect? I appreciate any thoughts on this one!
Ben
High model weights
2 posts
• Page 1 of 1
High model weights
by UTgrouse » Wed Nov 09, 2005 9:52 pm
- UTgrouse
- Posts: 1
- Joined: Wed Nov 09, 2005 7:23 pm
High model weights
by gwhite » Thu Nov 10, 2005 11:45 am
Ben:
Just be sure that the number of parameters was estimated correctly. For models that are overly parameterized, estimates often hit the boundary, and with the logit link, are not counted as being estimated when in fact they are.
Gary
Just be sure that the number of parameters was estimated correctly. For models that are overly parameterized, estimates often hit the boundary, and with the logit link, are not counted as being estimated when in fact they are.
Gary
- gwhite
- Posts: 340
- Joined: Fri May 16, 2003 9:05 am
2 posts
• Page 1 of 1
Who is online
Users browsing this forum: No registered users and 1 guest