In the hyper-competitive world of finance, companies are in a never-ending race to cash in on market intelligence.
Demand for innovation is simple -- a need for real-time access to market intelligence that allow them to be smarter and act faster than the rest of the pack. The revolutionizing induction first hit the street in summer 2005 as a technology spin-off to advance high performance grid computing (HPC) to a completely new stratosphere. Hence the birth of high throughput computing (HTC) on Wall Street.
While new for finance, these computing practices have been around for some time in government and research sectors. The old days of verbally relaying buy and sell orders from the sidewalks up to traders at the New York Stock Exchange Building on Wall Street are long gone.
Boutique technology firms have huddled in the financial meccas of lower Manhattan and London to design state-of-the-art trading platforms that drive some of today's most lucrative financial firms.
Notably, some of technology's greatest feats can be attributed to the financial industry as a whole. Hedge funds, brokerage strategies and buy-and-sell methodologies are all made possible through technology development and software implementation services. Investment firms contract technology groups who specialize in "the art of concept to code" that work under strict confidentiality agreements.
Working together, the world's greatest minds provide mathematical formulas, dynamic models, entity mapping, proofs and logic correlations that are built into functional tools to navigate across both public and private cross-trading networks.
To understand trading strategies you need to understand the basic principle of an algorithm. In short, it is an automated process by which a computer can make a decision based on pre-determined information (data) you feed it. It then performs the action based on the criteria you pre-set/program for the computer to make that decision.
Popular examples of algorithms are seen with Garry Kasparov, the world renowned Chess Master, who took on IBM's "Big Blue" computer -- and lost. Slot machines, aka the flashy one-armed bandit, use a random number generator (RNG) algorithm program. And as we know, the house always wins ... at least at slots.
Achieving success goes beyond having a fast computer. A computing center's primary responsibility is to provide high availability (computing grids) for complex event processing power needed to run 24/7 real-time analyses on multiple global trading platform environments.
Financial engineers, many with mathematics PhDs, have deep understandings of algebra and calculus as core ingredients to their investment modeling strategies. These numeric calculations are centric to the software logic design rules that result in trading profitability.
Depending on a fund manager's strategy, trading programs are made up of myriad historic and real-time information. Economic data, news updates, market conditions, risk analysis, liquidity pools, pricing schemes, investment instruments and exchange rates are among the data sets taken into consideration. Customized analytics are then built around these data sets in the form of applications, which allows traders running powerful algo trading programs to view and adjust market activity via advanced interfaces on their PCs.
Market information is then filtered against pre-selected "rules" that can identify fractions of a cent profit margins that trigger high volume investment actions -- e.g. "buy" or "sell" orders -- all within a couple milliseconds.
Flash orders are the host stage for algo trading to run trade matching functions to uncover opportunities in what I like to call a "liquefied resonance application server state" for millisecond transactional efficiency. The biggest industry sluggers are trading firm experts like Renaissance Technologies, Quantum Group and Goldman Sachs.
Fund managers and traders work together with customized smart market intelligence platforms that are rolled-out for day-to-day use. They monitor multiple trading exchanges worldwide searching for profitable opportunities.
Since 2005, HTC daily trade volumes have risen over 164 percent, per the NYSE. Technology has been the tool to allow the financial industry to access and digest dramatically higher volumes of data. It's fundamentally revolutionized daily trading activity. Known as high frequency traders, they target large volume transactions and represent a very small fraction of the overall trading population. They weed through millions of transactions to uncover opportunities to match "buy" and "sell" requests within dark liquidity pools for lower transaction fees in semi-transparency, as to not tip off their competition.
An estimated $21 billion in annual aggregate profits are generated from this low latency (grid computing) arbitrage investment strategy, according to TABB Group, a capital market research firm. It is important to note that a very small percentage of institutions leverage this technology.
Let's build one -- we can put the data center in the Caymans.
Ryan Peters is a technology journalist who can be found online at http://contactryan.wordpress.com
