Financial markets are valuation machines that are continuously pricing assets based on unending streams of information from every source imaginable. However, the signal to noise ratio is very low in these data streams. It takes some time for market participants to sort through this barrage of (mis)information and come to a consensus on price, however fleeting that equilibrium state might be.
Academic research and our own trading experience supports the view that, at the very least, markets are inefficient in the short term. In small time frames (seconds, hours and even days), trades tend not to be independent and identically distributed as economists would have us believe. Instead, they tend to be conditional and serially correlated. This leads to a common phenomenon of momentum trading.
Properly trained, intelligent systems can take advantage of such inefficiencies and can generate extraordinary returns over the long term. These small edges (more precisely, positive expectation trades) need to be sized correctly and given time for for the law of large numbers to apply and the returns be realized.
Computer systems can be trained to trade expertly, with discipline, and without fear or greed in the face of the chaos of the markets. They can sort and react to multiple, simultaneous market signals with superhuman speed and precision. And they can do this 24/7 without requiring a coffee or bathroom break.
Real-time, complex event handling
© Hedged Capital LLC. All Rights Reserved.