Best model

39 views
Skip to first unread message

Angela Fanelli

unread,
Sep 29, 2020, 2:15:52 AM9/29/20
to distance-sampling
 Hello,
I built several ddf models, 
What I am supposed to do if the best model (with lowest AIC) hasn't a good fit ( p value <0.05 when using the ddf.gof function ) ?
  Can I keep this predicted distance value or it is better to built another model? 
Do you have any suggestions for me ?  

Regards

Angela

Eric Rexstad

unread,
Sep 29, 2020, 2:55:01 AM9/29/20
to Angela Fanelli, distance-sampling

Angela

Setting aside AIC rank for the moment, making inference based upon a model that does not fit your observed data does not result in defensible estimates.  Your estimates will be open to criticism that you've used a model that is in conflict with the data you observed.

More reliable estimates are derived from employing models that fit your distance sampling data.

--
You received this message because you are subscribed to the Google Groups "distance-sampling" group.
To unsubscribe from this group and stop receiving emails from it, send an email to distance-sampl...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/distance-sampling/f8ec0fa3-ae65-4502-a9e6-fb8fa03713c7n%40googlegroups.com.
-- 
Eric Rexstad
Centre for Ecological and Environmental Modelling
University of St Andrews
St Andrews is a charity registered in Scotland SC013532

Angela Fanelli

unread,
Sep 30, 2020, 2:12:46 AM9/30/20
to Eric Rexstad, distance-sampling
 Hi!!!

I solved the goodness of fit problems with the the truncation of the most distant detections 

Regards
Angela
Reply all
Reply to author
Forward
0 new messages