Basics Of Automated Trading
The integration of technology throughout the financial industry has revolutionised the manner by which business is conducted. Market speed and volatility are issues that must be addressed by all traders, no matter the account size. In an attempt to adapt and flourish within the current electronic marketplace, traders and investors alike have chosen to implement automated trading systems within their portfolios.
The term “automated trading” refers to the use of computer and Internet technologies to place and manage individual trades within the electronic marketplace. The ability to automate trading practices fully makes it possible for a trader to implement his or her entire trading plan instantly, without having to interact with the market manually. It has been estimated that 50% of all trades executed on the equities and futures exchanges in the United States can be attributed to automated trading systems.1)Retrieved 8 February 2016 http://www.cftc.gov/idc/groups/public/@economicanalysis/documents/file/oce_automatedtrading.pdf
Advantages Of Automated Trading
Several advantages are afforded to the trader through the implementation of an automated trading approach. Automated trading (i.e. automated trading systems) can be used to execute complex trade management principles, reduce human error and define quantifiable “edges.”
The process of a trade’s execution consists of several elements, with each being essential to the success or failure of the trade. Automated trading provides opportunities for enhancement in these areas.
The first element of the trade is the trade’s entry, or order entry. In today’s electronic marketplace, order execution times are measured in milliseconds, with an actual profit or loss often realised in seconds. Automated trading provides instant order entry upon a trade setup’s identification. Without the need of trader discretion, the physical act of placing a trade upon the exchange is sped up exponentially.
The second element of trade execution is the trade’s real-time management. Automated trading enables the trader to instantly place a profit target and stop loss order immediately upon acceptance of the entry order into the marketplace. Trading platforms such as MetaTrader 4, NinjaTrader and Trading Station can be easily programmed to execute complex stop loss and profit target strategies. Scaling bracket orders, OCOs, and trailing stops are just a few money management “ATM” options included in the each of these platforms’ functionality.
Development Of Quantifiable Edges
Currently, electronic trading platforms available to the retail trader include robust functionality in the areas of historical data research and automated system development. MetaTrader 4, Trading Station and NinjaTrader offer diverse tools that can be used extensively for automated system development. Optimisation, walk forward analysis, and backtesting are just a few ways by which this evolving computing power can be put to use in an attempt to develop a quantifiable “edge” within the marketplace. Remember though, that past performance is no indication of future results as discussed below.
One of the most popular ways in which a statistical “edge” is developed by a trader is through a process called backtesting. Backtesting is the practice of applying a trading system to an older set of market data in order to measure its relevance.2)Retrieved 9 February 2016 https://wallstreetpilot.com/backtested-performance.html
The statistical results of backtesting can be of great use to the optimisation of a portfolio. Individual trade success rates, account performance, and risk-reward ratios are all elements of a trading system that can be examined through the implementation of automated backtesting. The ability for the retail trader to execute in-depth market research quickly and cost effectively is one of the major advantages made possible by automated trading.
Removal Of Human Error
As a trader interacts with the market, several challenges arise that are attributed to “human error.” Even though a trader may have developed a successful trading system based upon discretion, the trader’s execution of the system can come into question at any given time. Periods of drawdown or prolonged success can greatly affect a trader’s confidence and judgment. The use of an automated trading system can eliminate human emotion from executing trades based upon irrational decision making.
Automated trading provides the trader the ability to execute his or her trading plan with precision and consistency. Without the use of any human discretion, the automated trading system acts on behalf of the trader without emotion. A trade setup is defined, the according trade is placed upon the market, and either a profit or loss is realised. The trading system is executed in a precise and consistent manner, ensuring that the integrity of the system is preserved.
Disadvantages Of Automated Trading
Although the practice of automated trading is widespread and its influence is extensive among market participants, there are several large drawbacks that must be accounted for before a trader modifies the existing trading approach to become fully automated.
For an automated trading system to be a successful one, several key inputs act as prerequisites. Uninterrupted electrical access is needed to run the computers that execute the trader’s automated systems. Computer hardware used to operate the automated trading system must remain in proper working condition. Routers, dedicated hard drives and constant Internet connectivity are all required. Gaps or lag in electricity or Internet speeds can pose major problems to automated system performance.
Exchange-based hardware and Internet connectivity issues are rare, but possible. Natural disasters can shut down servers located at the exchange, and entire power grids can be vulnerable to interruption. In August 2003, Hurricane Isabel ravaged the northeastern portion of the United States and subsequently forced the shutdown of several power grids used to facilitate trading operations for the NYSE and NASDAQ exchanges. Both exchanges did have backup power systems and alternate order routing platforms in place, so the Securities Exchange Commission deemed that Hurricane Isabel “did not significantly alter financial markets.”3)Retrieved 10 February 2016 http://www.sec.gov/news/testimony/ts102003sec.htm
However, the interruption in power did have an impact on the ability of a client to access the exchanges. Therefore, if a client had an automated trading system actively trading instruments on the NYSE or NASDAQ, the client was very much at the mercy of how effective the backup systems of these exchanges were. Had the servers remained down for any length of time, open positions could not be managed and the result could have been disastrous to a trader’s portfolio.
Flawed Trading System Development
One of the key characteristics of an automated trading system is the methodology by which it was created. There is often a heavy reliance on the results derived from historical data analysis and backtesting to create a statistically quantifiable trading system. Although the ability to test a system’s validity upon historical data sets is a powerful tool, there are pitfalls that can cause the results of testing to be misleading and largely inaccurate.
Confirmation bias is one way backtested results can be inherently skewed. The term “confirmation bias” is defined as being the tendency to search for or interpret information in a way that confirms one’s preconceptions.4)Retrieved 10 February 2016 http://www.sciencedaily.com/terms/confirmation_bias.htm In many cases, traders subconsciously exclude pertinent data from a study, or craft the study in a fashion that guarantees a successful result. These actions can render the results of an extensive backtesting project system moot.
Another common pitfall in creating a statistically viable automated trading system is the unintentional backfitting of historical data. In many instances, automated trading systems are developed with maximum profitability being the primary objective. System optimisation, based on the analysis of historical data sets, aspires to properly align risk with reward to achieve maximum profitability. As the researcher alters profit targets and stop losses applied to the historical data, the system can become tailored to the historical data set. In essence, the automated trading system would be able to trade historical markets perfectly, and be next to useless in future markets.
Automated trading systems can be susceptible to software malfunctions, commonly referred to as “glitches,” located on the client side or at the exchange. A glitch is a small error or malfunction that can prohibit an entire operation from running smoothly.5)Retrieved 10 February 2016 http://www.businessdictionary.com/definition/glitch.html As it pertains to electronic trading, a glitch is a minor error in the programming of an individual trader’s automated trading system, or a programming error in the order matching software at the exchange. Although a glitch may be a minute discrepancy, the possible impact upon market participants can be catastrophic.
In 2013, investment banking firm Goldman Sachs experienced a software glitch that caused an onslaught of unwarranted trading activity on Goldman’s behalf. The resulting loss was estimated to be in the hundreds of millions of dollars.6)Retrieved 8 February 2016 http://www.computerworlduk.com/news/applications/goldman-sachs-hit-by-trading-software-glitch-3465034/ Albeit on a large scale, Goldman’s glitch is a good illustration of how an obscure programming error in an automated trading system can cause trading operations to run wild.
The speed of electronic markets demands that a trader is efficient in nearly every aspect of his or her trading. Automated trading offers many advantages to the trader, ranging from order-entry speed and precision, to the implementation of a comprehensive trading plan. However, risks do befall traders who are exclusive to automated trading. Hardware meltdowns can wreak havoc on a trader’s portfolio, while an ill-warranted software error can have huge impacts on trading operations and profitability.
A solid trading decision is one where reward outweighs risk, and the pros outweigh the cons. If the need for computerised precision and speed is an integral component of your trading approach, then automated trading is more friend than foe. If your trading methodology is rooted in discretion, and speed is not a crucial factor, then automated trading systems are likely an unnecessary undertaking. Ultimately, the decision of whether or not to automate lies with the trader.
Any opinions, news, research, analyses, prices, other information, or links to third-party sites are provided as general market commentary and do not constitute investment advice. FXCM will not accept liability for any loss or damage including, without limitation, to any loss of profit which may arise directly or indirectly from use of or reliance on such information.
References [ + ]
|1.||↑||Retrieved 8 February 2016 http://www.cftc.gov/idc/groups/public/@economicanalysis/documents/file/oce_automatedtrading.pdf|
|2.||↑||Retrieved 9 February 2016 https://wallstreetpilot.com/backtested-performance.html|
|3.||↑||Retrieved 10 February 2016 http://www.sec.gov/news/testimony/ts102003sec.htm|
|4.||↑||Retrieved 10 February 2016 http://www.sciencedaily.com/terms/confirmation_bias.htm|
|5.||↑||Retrieved 10 February 2016 http://www.businessdictionary.com/definition/glitch.html|
|6.||↑||Retrieved 8 February 2016 http://www.computerworlduk.com/news/applications/goldman-sachs-hit-by-trading-software-glitch-3465034/|