Black-Scholes...

The Black-Scholes equation is little-known outside of financial circles.  It's challenging to explain if you don't understand the financial instruments (options) that it deals with, and I'm not going to try to explain them here (but the Guardian has a readable article).  I want to make a different point: that the Black-Scholes equation is, essentially, a mathematical model.  It is universally implemented on a computer, so it's fair to call it a “computer model”.

The Black-Schole computer model is a very simple example of such a model, based on a single equation modeling an isolated financial instrument.  The data underlying the model are as perfect as is possible to obtain.  Yet the financial industry – even those who understood the mathematics – badly misused it.  As best I can tell, the misuse arose entirely because of the lack of a better alternative.

The computer models being used to investigate climate change, by contrast, are enormously complex – many of the models contain hundreds of equations and millions of interconnected elements being modeled.  The data underlying the models is known to be full of systematic errors, unexplained negative correlations, and missing large pieces.  No single human fully understands all the mathematics involved. 

Without even looking more deeply than this, I cannot fathom why so many scientists are willing to bet their careers on those climate models...

Trouble with Models...

An excellent article about the dangers of relying on computer models to predict future events.  They had me at the introductory quote, one of my old favorites:

“Prediction is very difficult, especially about the future.”
—Niels Bohr, Danish physicist (1885 – 1962)