Advantages of Software Automated Trading Strategies
Data, Discipline, and Rigor
Automated trading strategies use scientific methodology to counter human cognitive and emotional biases and generate superior investment decision-making. Quantitative models have been proven to capitalize on market inefficiencies and have the potential to earn superior profits.
Automated trading strategies provide numerous advantages for implementing successful investment strategies. A rigorous and disciplined approach can lead to profitable strategies far superior to human discretionary trading.
Automated trading is disciplined trading. The strategy will do exactly as the underlying software is written. The software will enter trades based on the core logic of the strategy and likewise exit trades according to its exit logic. Irrational human behavior and biased decision-making do not interfere.
A well-defined strategy is;
(1) back-tested on a reasonable set of data; and
(2) forward tested for a reasonable amount of time.
These two elements are the beginning of a good strategy. The amount of time or data quantity to test over is another topic that deserves its own article. For this article, the scope of the discussion will be limited to the high-level key factors of automated strategies.
Forward testing consists of real-time streaming data, but the trades are taken in a sim account. This approach helps find bugs in the mechanics of the strategy while using real-time streaming data. Think of this as the validation phase of the software, where execution is tested across various paths of software. Strategy development follows the traditional path of software development, validation, and bug fixes.
Once confidence has been gained in forward-testing then it is;
(3) tested in a live account with real money using a small number of shares, contracts, or pips.
In this phase of development, the objective is to test the communication between the exchange, test margin requirements (CME changes them based on volatility) and most importantly ensure trades are filled as expected. The number of buyers and sellers and the corresponding price determine the fills.
Once a strategy works, it can be scaled to trade more shares, contracts, and pips.
(4) The ability for the same strategy to scale across multiple markets and multiple instruments within that market is paramount.
Discipline is key. Follow your system and trust your research.
The software trades exactly as designed.
(5) No human bias,
(6) No knee-jerk reactions,
(7) and no sentiment trading.
Software does not have sentiment (though there have been recent discussions around AI having feelings, we’ll ignore possible sentience here :), therefore there is no attachment to any market. However, humans do have an attachment and we’ve all been there. We have a stock or two that we love to trade even if we have lost money with it. So here is another reason automated strategies are superior to human trading.
(8) Automated strategies can comprehend numerous indicators and mounds of data in real-time to make a decision.
Trust the model.
Humans do not have the mental capacity to watch multiple markets and trade in all of them simultaneously, but automated strategies can. At Arcadia we prefer to use basic and simple strategies, however, we do not shy away from complexity and will use the most advanced tools in technology to make the right trade decisions. As Einstein said, “make things as simple as possible, but no simpler.”