> The parameter space is immense This. This. This. I've seen claims that the number of parameters in the Imperial College agent-based model was c.a. 400. Perh

> The parameter space is immense This. This. This. I've seen claims that the number of parameters in the Imperial College agent-based model was c.a. 400. Perhaps it was 40. Compare to a quadratic or quartic fit to some log-lin data. If you have to represent your ratio of poorly-constrained model parameters (40 or 400) to apparently necessary parameters (say 4) _using Big-O notation_, then science has arguably left the premises.