Parameter settings for the SAM assessment model need to be settled on.
Propose to base the approach on a series of scans where we bind parameters
together and then test via AIC or LR tests whether the reduced model is
significantly different from the parent model.
Parameters to test in this manner:
1. Catchability-at-age and observation variances of HERAS survey
2. Catchability-at-age and observation varainces of IBTS-Q1 survey
3. Catchability-at-age and observation variances of IBTS-Q3 survey
4. Fishing selectivity-at-age and observation variances.
Other models to test
1. MLAI/SCAI power-law vs linear model
Original issue reported on code.google.com by markpayn...@gmail.com on 25 Nov 2011 at 3:14
Original issue reported on code.google.com by
markpayn...@gmail.com
on 25 Nov 2011 at 3:14