I find it odd for people to actually forecast real economic events with mere econometric tools. I mean even if you can come up with thousands of variables to put into your regression model, you could never account for all the uncertain variables exist in real life.
Mathematical models are elegant, accurate, but reality is not so precise. It is full with random events and shocks that no human mind or computer can ever think of.
But still, a mathematical model is a good approximation of the reality. You can’t accurately forecast, so you do your best to guess.
I wonder if it is indeed possible to create a new mathematical model, taking into account chaotic situations and all permutations that life could present.
Indeed, that would be a beast. :D