BiologicalRecordsCentre / sparta

Species Presence/Absence R Trends Analyses
http://biologicalrecordscentre.github.io/sparta/index.html
MIT License
21 stars 24 forks source link

Add option to fix alpha.p across years #99

Open JKSheard opened 5 years ago

JKSheard commented 5 years ago

Find a way to loose temporal element to alpha.p, allowing it to be fixed across years. @drnickisaac for Friday

drnickisaac commented 5 years ago

It's trivial to replace basemodel <- "logit(p[j]) <- alpha.p[Year[j]]" with basemodel <- "logit(p[j]) <- alpha.p" in getObsModel.R. In theory it would be easy to change the prior from: for (t in 1:nyear) { alpha.p[t] ~ dnorm(eta.p0, tau.lp) } To alpha.p <- eta.p0 with tau.lp deprecated. The difficulty is that the priors remain in a text file, rather than being meta-coded. So this is actually dependent on a much larger job of meta-coding the priors.

drnickisaac commented 5 years ago

Clearly this is dependent on issue #108 in order to be tackled