Hi all,
I'm working on an ODE model where I need to infer a parameter μ and a non-parametric function φ(t), and I'm looking for recommendations for tools or approaches in R to solve this problem. My dataset is small, involving just 7 data points in the form (t, n).
The model is given by:
dn/dt = φ(t) - μ * n(t), n(0) = 0
It has an analytical solution in integral form:
n(t) = ∫0t exp(-μ (t-s)) φ(s) ds
I've been exploring the fungp and magi R packages, as they both use Gaussian processes to model φ(t). However, I’ve encountered a couple of challenges:
- fungp doesn’t seem to support parameter fitting, which is crucial for estimating μ.
- magi doesn't allow for parameters that vary over time (which is the case for φ(t)).
Can anyone suggest other R packages or approaches that might be a better fit for this problem? I’m particularly interested in methods that can handle both the parameter fitting for μ and the time-varying nature of φ(t).
Thanks in advance for your help!