Numerical Weather Prediction is a field where meteorologists produce weather forecasts through large computer simulations by taking the classical physical laws governing the flow of the atmosphere, expressing them in mathematical equations, and then solving these equations using high-performance supercomputers. (See Post #626.)
When they do, they don't just produce a single forecast. Rather, they produce multiple forecasts in what's known as an
ensemble prediction system, the idea of which is to take account of Chaos Theory (trying to represent all of the little uncertainties that exist in the atmosphere when doing numerical weather simulations).
For this reason, the best practice is to run several simulations of the weather model so that they can get a handle on the uncertainty in the atmosphere to get a probabilistic output.
However, because I don't have access to incredibly powerful supercomputers to deal with the huge volumes of data I would ideally be ingesting and analyzing, I have to adjust my
Numerical Price Prediction forecasts on the fly, in real time, based on the "laws" governing the flow of asset prices.