AN OVERVIEW OF ALGORITHMIC TRADING

Trading in securities, derivatives, futures, and other financial instruments invokes images of trading floors filled with hundreds of people shouting, milling about, or peering intently at computer screens. The reality is that today, algorithms running on computer servers conduct more than 70 percent of trades in the United States. In the financial industry, speed matters. The company that can first detect and act upon an opportunity stands to profit handsomely. For some trading strategies, getting in the queue first wins the trade. No wonder, then, that companies go to great lengths to ensure they have an edge — even if just a microsecond — over the competition.

In an algorithmic trading system (ATS), algorithms running on high-performance computers handle trades and make decisions. Such systems use multiple techniques to minimize latency and gain a competitive edge. These techniques include co-location with the different exchanges, use of shortest path networks, and, increasingly, utilizing 10GE to achieve lower latency. The major trading companies are adopting all of these techniques to gain an advantage over the competition