28 Mar Systematic Trading
Systematic Trading is the selection and execution of trades based on a predetermined set of rules generally drawn from a body of work known as technical analysis. In futures markets, such analysis utilizes a data base that includes market price, volume and open interest information. In the case of stocks data could include fundamental performance metrics such as earnings, cash flow, outstanding shares, and short interest.
Modern Systematic trading features computer automation of analysis and data, largely originating in the late 1980’s. Prior to the proliferation of personal computers, traders performed calculations by hand employing charts and related data.
Today systematic trading primarily relies on the price data generated by the markets and provided by exchanges. Significantly, most systematic traders believe that all market information is reflected in the market price.
Why should we embrace systematic trading?
All successful traders have one thing in common – they all have rules that govern their trading. It is every trader’s mission to cut losses and to let profits run – yet few are able to do it due to emotion and ego. The stress of a chaotic market especially requires a predetermined set of answers. Success demands adherence to a well researched system of rules. Systems can of course be changed, but not during a trade! Change it, alter it abandon it for something different – all are viable. But not when in a trade! Just do it … until you find something better.
Rules are the enemy of emotion and ego. Many people make rules but they have a tough time following them. Following a bad set of rules is invariably better than having no rules at all. It was once told to me by a smart man “why would you spend all that time researching a trading system only not to follow it once you built it?” The reason? Because my ego and emotion told me not too! Somehow all that research was worthless once a trade went against me and I had to change it midstream. If anyone has done this, and we all have, you know what I’m talking about!
How to choose a system
We’ve all heard that hind sight is 20/20. Many trading systems show historical performance that is very pleasing to the eye in the form of a graph or to a sort in the form of Sharpe (Sharpe is reward/risk) so should we choose a trading system based on how it has performed in the past?
The answer is not so easy. Many variables need to be considered. For example, an assessment as to how the system performed historically when market conditions were as they are currently would be nice. In times of market stress did the system perform well? Are we in those kinds of times now? Volatility seems to be something that is important to systems. How does a system perform in high volatility or low? What was the market doing during these times?
The above are all good and certainly reasonable questions. However, if we found that a system was great in a low volatility environment and bad in a high volatility one would shutting it off during the latter help the performance? When would we do that? How would we do that? Before you try to answer, let me point out that system designers have been trying to figure this out for some time – to no avail. Generally speaking, there is no direct relationship between volatility and performance. We all know it’s there but what is it?
For professionals this is difficult to discover. For most people this is out of the question. For the person using a system based approach to trading delving into the internal dynamics of systems is all but impossible. A tweak here will cause a change somewhere else. Continuous tweaking often increases the “degrees of freedom” suggesting curve fitting. So what so we do if we have all these systems and symbols to choose from? The answer is really quite simple. Choose many of them!
Most people know that the best approach to trading or investing successfully is through a portfolio approach. This is well established. Things don’t always work the way you want so to smooth it out we simply make smaller best in many things, all with positive expectations. When we choose a set of systems against symbols we should be mindful that even though the past doesn’t predict the future it does lend itself at least to best guess at direction! After all we would not trade a system on a symbol that consistently lost money over time.