By Geoffrey Lauprete
An alpha is a model that predicts the prices of financial instruments. And while the idea of modeling the markets and predicting prices was not new back in the 1980s and 1990s, it was during that era that cheap computing power became a reality, making possible both (1) computational modeling on Wall Street trading desks, and (2) the generation and collection of data at a rate that is still growing exponentially as of the writing of this chapter. As computers and systematic data collection became ubiquitous, the need for innovative modeling techniques that could use these newly-created data became one of the drivers of the migration of PhDs to Wall Street. Finally, it was in this climate of technology evolution and exponential data production that the quantitative trading industry was born.
Quantitative trading and alpha research took off at the same time that cheap computational power became available on Wall Street. Alphas are predictions that are used as inputs in quantitative trading. Another way of putting it is to say that quantitative trading is the monetization of the alphas. Note that an alpha, as a form of prediction model, is not the same thing as a pure arbitrage. Sometimes the term statistical arbitrage is used to describe quantitative trading that exploits alphas.
Note that one could debate whether alphas ought to exist at all – some of the arguments ...