Time varying input for ODE solving (aka (?) forced ODE)

[miquel_palmer]

I am still balancing the learning effort and the new capabilities of STAN. The build-in integrator and the possibility of estimating parameters from ODE is certainly interesting. I have playing with the predator-prey model in

https://mc-stan.org/users/documentation/case-studies/lotka-volterra-predator-prey.html

This example is very clear. Congratulations.

Toward the end, at “Exercises and Extensions #8”, it is proposed to evaluate the effects of external “covariates”, as Temperature, which typically change at smaller scales that the process of interest, are densely sampled and remains unaffected by the process. Thus, the example in

Forced ODEs, a start for a case study?

Hi! Attached is a small intro to forced ODEs in Stan. I am not sure if this falls into the category of a case study (no data is fit at all)… one would probably have to add the standard example of the Theophylline data set or something similar, not sure. In any case, this is the type of info which I see users on the mailinglist ask from time to time. Therefore, we should consider to put it up somewhere if people agree that this document is helpful. Suggestions and improvements are certainly ver…

seems not to be the same case (…and it is hard to understand for me…).

Can someone provide a detailed (=beginner’s level) example of how to deal with this type of external variables?

For those using R and the deSolve library, I am looking for something that emulates aproxfunc.

Thanks
miquel