search

curryc2 / AFSC_VAST_Evaluation

Repository for evaluation of VAST package for single and multispecies spatial delta-generalized linear mixed models (delta-GLMM). The purpose of this analysis is to evaluate differences in indices and uncertainty estimates generated by VAST and design-based estimators, for a range of species in the Bering Sea and Gulf of Alaska.
1 stars 2 forks source link

Design based estimates #3

Open GreatDana opened 7 years ago

GreatDana commented 7 years ago

For these comparisons it may be helpful to exclude 2001 from the Gulf of Alaska design-based estimates as the entire eastern GOA was not sampled. Its good to use it in VAST though because it fills in the missing areas/depth strata.

curryc2 commented 7 years ago

Excellent suggestion. I can either: 1) Forego including 2001 in all future figures/summary comparisons, or 2) Add function argument for years to calc design-based index. Preference?

GreatDana commented 7 years ago

Since its sort of special case for the GOA in one year, probably just exclude 2001 from the design-based indices, unless a year function might be useful for testing how many survey years make VAST useful? Like is it any good if you just fit 2011-2015 for example?

Cheers, Dana

On Mon, Apr 10, 2017 at 12:26 PM, Curry J. Cunningham < notifications@github.com> wrote:

Excellent suggestion. I can either: 1) Forego including 2001 in all future figures/summary comparisons, or 2) Add function argument for years to calc design-based index. Preference?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/curryc2/AFSC_VAST_Evaluation/issues/3#issuecomment-293068541, or mute the thread https://github.com/notifications/unsubscribe-auth/AGUGxt6ubeWxmVtsTDJvHKRgRXCfGPXfks5rupB9gaJpZM4M5Qti .