An estimated model fit for a time-series model.

An equation that may require fixed or free parameters, the current time t, the current response value, and / or external environmental time series.

Computes the hidden state’s initial value at t0 from fitted parameters and baseline known values.

The negative log likelihood given the predicted and observed per-variable time-series and a function to retrieve parameters required by the likelihood function.

A function that computes a measured system property given a current (time t) and previous (time t-1) system state.

The model system may make use of either stepped or integrated-time equations. Each are both require the same inputs, but differ in whether a solver is applied or not. Here, the DU's purpose is for additional type safety. TODO Allow state to differ between equations.

A function that returns a parameter's current value by its name.

Predicted time-series for a single variable. 'u is the state unit of measure.

A single variable’s observed vs expected values after model-fitting.