With the advent of electronic exchanges, liquid markets have become a confusing cocktail of day-traders, institutional investors, fundamental producers and consumers, and high frequency algorithms. For even the most experienced trader, the complexity of modern markets represents a difficult puzzle. Price movements are now shaped by an unprecedented amount of real-time information which must be continually processed.
In every sense, trading is a game. There is directly accessible fundamental information. There is important indirect information that must be handicapped. There are very specific rules of execution. And, most importantly, there is a score (ROI) to be kept. Despite all of the technological advances in trading, information is still processed through a human (or human directed) filter. Ultimately, it is the behavior of traders stimulated by data that dictates price movements.
Modern computing power can enable this game to be played at a very high speed with an enormous depth of data. Our methodology incorporates a “clean slate” analysis of plausibly causative factors in combination with advanced statistical, mathematical, machine learning and behavioral modeling techniques.
Powerful tools require equally powerful discipline. Cognitive biases can disrupt the accuracy of apparently successful trading models. Through our multi-tiered validation approach, we can accurately assess the effects of randomness and provide compelling evidence of potential edge before the implementation of live trading.