Automatic Way Forward

Today's markets pose significant challenges to those seeking trading profits; despite the volatility experienced over the last couple of years. In contrast, markets of 20 years ago held a plethora of trading opportunities, often obtainable from trading just one product or even one stock. So why have things changed so much? Quite simply we have more market participants embracing ever changing technology resulting in a more mature market where opportunities are not so apparent and need to be weeded out and traded with the ultimate precision and efficiency.

That's not to say that opportunities don't exist in abundance in some of the emerging marketplaces; but here exists a notoriously volatile and illiquid climate, often taking control away from the trader.

To maintain a degree of control, we need to focus on the more mature markets, where our trading can follow a pattern ensuring swift and accurate execution of our directional orders.

So how do we find these opportunities and how do we trade them?

Strategy & Signal Generation

Firstly we need to define our trading strategy. This ultimately means identifying what set of circumstances result in us deciding to enter and exit our trading position.
We could spend hours manually scouring charts or fundamental reports looking for instances where our strategy criteria fit. But that is unrealistic, so we need to utilise the scanning technology available. If we have a predefined strategy and an automated means for searching for products where our trading criteria fit, then we are half way there. We all know that to have a strategy is crucial for successful trading; however the means by which we execute that strategy is of much greater importance. It is fair to say that you can succeed with an average strategy and excellent discipline but not with an excellent strategy and average discipline. So we need to have a plan, but more importantly, an effective means of executing that plan.

When we have our plan or strategy we can code the rules into a software programme, which in turn analyses market data, and generates our trading signals based upon our predefined trading criteria. Here our trading system is born.

Back-Testing & Optimisation

Fortunately we do not need to be computer programmers to be able to turn our rules into code, as many software programmes have easy to understand language and wizards to take you through the process, step by step.
Having programmed our rules into the software, we can back-test these rules on historical data, to see how it would have performed in the past. This is by no means a guarantee of how successful your strategy will be going forward, but it will give you an indication as to the potential of your trading strategy.
By reviewing the back-test performance results, a trader can optimize the variables of his system to refine the profitability. Once again, some of this optimisation can be automated with the software itself identifying which variables would increase the historical profitability.

Trading the Strategy - Simulation Vs Live

We can take our development to the next level since some software programmes allow you to then test your strategy in a simulation environment. Allowing you to trade the market without risking any money is the ‘icing on the cake' as far as testing a system before going live. However for a few reasons there is never a substitute for actually trading the market. Trading factors like slippage and part fills, which often occur in the real world, can have a significant effect on a system's performance. Let's face it, if this automation process of developing a trading system resulted in a guaranteed money making machine, then we would all be sitting back counting the money.

It is often the way we execute the signals that our system generates that makes the difference between success and failure. We all know that discipline, or more accurately a lack of discipline, is probably the number one reason some traders fail to make money and after all, we are human and prone to emotion, which is more often than not the catalyst for poor and irrational trading decisions. The problem is that discipline is not always easily learnt, and certainly not overnight!

However there is an overnight solution to adhering to a trading strategy and that is ‘Automated Execution'. It is incredible that even when our trading system is shouting ‘buy' or ‘sell' we still sometimes struggle to pull the trigger as our emotions on what we ‘think' is happening in the market take over and result in us avoiding commands to trade.

By automating our execution too we have developed a true front to back automated trading process.

Automated Execution

‘Automated Trading' per se can now be described as the identification of strategy criteria, signal generation and execution of that defined strategy... all automated.

It sounds like we are just generating a computer to do all the work. Well, to a certain extent we are, but the markets are forever changing and we need to have some flexibility in what we do to accommodate those changes.

However some strategies have their logic ‘locked' away, so that individuals can use or subscribe to those signal systems without full view of the thoughts processes and methodology that are behind them.

Black & White Boxes

A ‘Black Box' is a term used to describe a ‘closed' or ‘sealed' trading system. By this we mean an algorithmic strategy which generates signals in any chosen market; but which cannot be changed or adjusted. It is then possible to have the trading signals executed automatically, without you ever lifting a finger. This is perfect if the Black Box strategy is a good one, and some of the good ones may stay profitable for more than 5 years, but many start out as winners and soon loose their inherent effectiveness and cease to be profitable...very quickly.

An alternative, which is becoming increasingly popular, especially amongst trading institutions, is the ‘White Box' system approach.

A ‘White Box' system is subtly different to the Black Box, in that it is your personal algorithms that make up the strategy. The advantages of this approach is the ability to easily tweak or adapt the system to changing environments or market circumstances; whereas a Black Box is closed and does not allow any adaptation to the ‘factory fitted' capabilities.

Algorithmic trading, in particular using the White Box approach is rife in the hyper- competitive trading world, as financial institutions now feel the mounting need for technology that aids their unique trading style. With the White Box approach it is possible to rapidly compose or evolve algorithms to monitor, analyse and respond to market events in a specific way.

Although the pre-defined Black Box algorithmic strategies are a great place to learn the ins and outs of automated trading systems, the user defined White Box approach had greater flexibility and therefore greater longevity.

The advantages of automated trading

The automated execution has the head line advantage of removing you, or should I say your emotions from the final decision of whether or not to push the trade button. If you believe in your strategy then you should trade it whatever the market feels like at the time of signal generation. To have the link between signal generation and trade execution taken care of by some automated process, means that discipline is maintained even during the most volatile market phases.

• The transparency of most markets means that the edge is harder to find. Automated algorithmic systems can scan multiple markets looking for a criteria fit.
• Trading psychology and discipline are difficult to master. Automatic execution removes the emotion from trading, making it easier to follow your strategy precisely.

The disadvantages of automated trading

Nothing is guaranteed when it comes to trading the markets and guaranteed profits with automated trading systems is no exception.

• If your trading system is not properly coded and tested, then profitability is unlikely.
• Sometimes it is impossible to put certain rules into code, which makes it difficult to develop an automated trading system.

As trading technology continually advances, the race is on to find algorithmic supremacy. As a result some strategies are becoming forever more complex, demanding a greater need for automation. Now that the execution of trading strategies can be automated, the risks of ill discipline and human irrationality have all but been removed.

Simon Brown
Managing Director
ProSpreads
www.prospreads.com