Closed seananderson closed 5 years ago
Could be a neat little model to package up given that these calculations are moderately popular and everyone calculates them without uncertainty.
Great! Thanks Sean. I'll do a double check on the code and mull over packaging these, but doing so would probably require a few more conversations about what's actually happening under the hood. You're right that they are gaining in popularity though!
Use the TMB one (
synchronyRetro-window-tmb.R
), not Stan. There are problems with the Stan one that I haven't resolved.Please check that I haven't made a mistake via R code — especially for CV_c and CV_s. Phi calculation looks right to me.