Partition Uncertainty for Model Forecasts

[jzwart]

Using this NEFI exercise as an example, let's try to estimate the contribution of initial conditions, parameters, drivers, process, and random effects uncertainty to total model forecast uncertainty.

[jzwart]

• [ ] Make a function to call different deterministic functions like this random walk example
• [ ] Iterate through days of year and run uncertainty partitioning code to see how relative uncertainty changes based on forecast date start

[jabrent]

Good ideas Jake! I can try to generalize the deterministic function part to be able to call other models and I'll loop through the uncertainty partitioning code to see how it changes with forecast date start and maybe year and/or site?

[jzwart]

That'd be great, thanks @jabrent !