We also considered non-uniform sampling, such as by sampling certain parameters in the log space, but it was challenging to identify which methods were appropriate for each parameter. Thus we leave this to future work.

Data Fitting: Optimization

While reasonable defaults for parameters may be helpful in uncovering the general dynamics and behavior of the model, of particular interest is deriving parameter values that result in model output that matches the real world. In order to do this, we attempted to fit our new model against real world data using numerical optimization methods.

Real-world Data Sources