Machine Learning and Trading: Fina Technologies

Company applies machine learning to quantitative trading. It is a spin-off of Gene Network Sciences (GNS) of Cambridge, Mass.

http://www.finatechnologies.com/about.html

CEO is Joshua Holden

"... trading expertise covers US Government Bonds and Options, US Agency Debt, FX spot and forwards, and US$ Derivatives (Swaps and Volatility). He has held desk-head positions at Goldman Sachs, Deutsche Bank, and most recently Countrywide Capital Markets. At every stop, he has focused on applying cutting-edge technology to the problems of price & model discovery, execution, and risk-management. Josh graduated MIT in 1993 with both a BS and MS in Electrical Engineering."

Investors include Reed Elsevier Ventures; spinoff from Gene Network Sciences,

An article by Joshua Holden appears in Forbes, "Why Computers Can't--Yet--Beat The Market"
http://www.forbes.com/2009/06/18/fina-financial-markets-opinions-contributors-artificial-intelligence-09-joshua-holden_print.html

"Training a financial trading system to deliver the single best model given the data will most often lead to models that fit the past at the expense of predicting the future. What are needed are systems that are flexible, adapt to changing circumstances and are, at their core, probabilistic rather than deterministic. By having distributions of possible models for the state of the world, we can balance the competing desires for certainty and flexibility. By retraining the models automatically when results begin to drift relative to expectations, we can achieve some of the adaptability that humans exhibit in the face of shifts. The path to beating the markets lies in building systems that understand, but do not emulate, the persistent biases in human nature."

"If our goal is to build intelligent systems to beat the markets, we cannot simply ignore irrationality. As Keynes famously remarked, "the markets can remain irrational for longer than you can remain solvent." Longer, too, than can an AI trading system."

In Forbes, "Man vs. Machine on Wall Street" http://www.forbes.com/2008/11/22/supercomputers-biology-quants-biz-wall-cx_mh_1124quants_print.html

Origin of the applying machine learning to finance: "The idea comes out of systems, or network, biology. Genes and proteins interconnect in a complex web. By drawing these connections, companies hope to invent better drugs. Merck in particular has put technology similar to that used by GNS at the center of its approach.

This computerized approach to biology attracted investors who were, in some cases, quants. Two years ago, Hill was having drinks at an upscale Manhattan bar with an investor and a GNS board member named Thomas Paul, then chief investment officer for Fortress Investment Group. For years, they'd been toying with the idea of using GNS' technology to trade stocks. That night, for some reason, the idea finally stuck.

Paul graduated from MIT in 1993 with bachelor's and master's degrees in engineering and computer science, and, like many of his peers, went to Wall Street, working first at Goldman Sachs and then at Deutsche Bank before starting an $800 million fund at Fortress.

He was prepared for the odd world of quants, he said, by playing on the MIT blackjack team--a different version of the team portrayed in the movie 21, in which a group of students figured out that with investor backing, they could consistently beat the house in Las Vegas by counting cards."

Founder of GNS and Fina is Colin Hill: "Colin Hill brings years of hands-on scientific experience to his role, with expertise in the areas of computational physics and systems biology. Hill is a frequent speaker at international scientific and industry conferences and has appeared in numerous publications and television segments including The Wall Street Journal, CNBC Morning Call, Nature, Forbes, Wired, and the Economist. He also serves as a board member of AesRx, a biopharmaceutical company dedicated to the development of a new treatment for sickle cell disease (http://www.aesrx.com). In 2004, Hill was named to MIT Technology Review's TR100 list of the top innovators in the world under the age of 35. He graduated from Virginia Tech with a degree in physics and earned master's degrees in physics from McGill University and Cornell University."