Automated trading software can be classed according to the market it trades or according to the type of user. The general market types are stocks, commodities, and currency. Users are either individual traders or large trading funds. Individual traders often use retail software, and funds develop their own.
Funds trade stocks, currency, and commodities futures using algorithms they have developed and incorporated into their automated trading software. The exact mechanics of these algorithms is carefully guarded. Stock trading funds frequently use so-called momentum algorithms, often combined with pairs trading. Momentum algorithms capitalize on short term, high volume price changes. Pairs trading means selecting two stocks in the same sector, then buying the stronger and shorting the weaker.
Currency trading funds are apt to use algorithms based on detection of price and volatility expansion in one direction. Price distribution characteristics of the currency market allow this approach to be successful, whereas it is marginal in stock trading. Momentum trading is used in some automated trading software that funds use to trade grain futures. At least one fund uses the physics of fluid dynamics to trade stock market futures.
Retail software, in contrast, tends to offer a pretty appearance on the computer screen, combined with the ability to place several orders simultaneously with the press of a screen button, and claim that this is “automated” trading software. The trader is still required to monitor the market, decide what stocks, futures, or currencies belong in his button-traded-basket, and then decide when to enter the trade. Some retail software offers a coding interface that the trader can use to write and test his own algorithms.
Another variety of retail automated trading software is referred to as black-box trading. The algorithms used by the software are not disclosed to the purchaser, and often he is not provided a suitable opportunity to test whether the software is actually profitable. Purchasers are asked to accept on faith that the software has been profitable in the past and will be in the future. Black-box trading systems are offered for trading stocks, commodities futures, and currency.
Funds that use automated trading software test it extensively. They use both bootstrap and Monte Carlo testing to see whether the system is profitable in the lab. When they find algorithms that succeed in their testing regimen, they test it in real time. Fund generated software controls all aspects of trading: it determines how much to trade, what to trade, and when to trade, and it places the orders directly on the exchange. An individual trader needs to emulate the testing strategy of the funds if he is to have reasonable prospects of trading profitably.