nicholasjclark / mvgam

{mvgam} R 📦 to fit Dynamic Bayesian Generalized Additive Models for time series analysis and forecasting
https://nicholasjclark.github.io/mvgam/
Other
91 stars 11 forks source link

Logistic growth with user-specified carrying capacity #31

Closed nicholasjclark closed 7 months ago

nicholasjclark commented 7 months ago

Sample code from the prophet paper can be used to build logistic trends with potential changepoints: https://github.com/facebook/prophet/blob/main/R/inst/stan/prophet.stan. This requires a carrying capacity cap, which is on the scale of the outcome. Would likely need to take the log of trend for log-link models so it could be directly included in the likelihood functions, such as poisson_log_glm.

nicholasjclark commented 7 months ago

All done now