Modelers were the whipping boys of the financial crisis – no news there – and there were more than a few who thought the financial world would be better off without them.
Enter Nick Albicelli, an FSA and quant for a Wall Street investment bank. He offered up this on an ERM listserv I follow:
[E]veryone just wants to point out how relying on models got us into this mess (take your pick – financial crisis, BP oil spill, etc.). Except for merely rote actions, everyone uses models for decision-making whether they claim to or not because everyone simplifies the world to deal with a manageable amount of information – that is the definition of a model.
When someone is arguing against using ‘models’ . . . they are really arguing against using a certain level of math in their models. But heuristic models, and models based on experience, with no formulas or computers or data are still models, so the question is not whether or not to use a model, but rather which model will give the ‘best’ results, for some definition of ‘best.’
Nice – it really focuses on the actual choice – which would you prefer:
- A model where a team of people collect data, sift through it and follow a rigorous process to a logical conclusion, or
- A model where you call up a coupla board members and do whatever comes to mind?