Design of the trading system

This paper is devoted to trading systems. Let us begin with a discussion of principles that should be taken into account when designing a successful trading system.

Emotions
Mechanical systems do not have emotions or intuition, which may be both advantages and disadvantages. Perhaps the control of emotions is the main reason why traders want to find a successful trading system. How many times have you mentally reproached himself that is not traded on a signal or image patterns, when they noticed? Psychological the tug of war between fear and greed can be quite debilitating. You want to give up "pressing shutter" to carry out trade in your computer.

Emotions and intuition can not be calculated mathematically, and mathematics - so everything should work the trading system. So the first and most fundamental principle in designing trading systems - is that every rule or reason to enter the market and leave it to be mathematically justified.

Complexity
The second principle is "KISS" - approach. (KISS - keep it stupid simple - means "keep it simple and stupid"). The natural tendency is to make the trading system is too complicated. The system, which initially began with a few rules, too often becomes too complex when adding more rules to include or exclude specific situation. You must be mentally able to understand and remember the rules. If you can not remember all the rules, and immediately identify the cause for alarm, the system is too complex.

Trading system grows in a complex with the addition of rules to accommodate more weight by blocking a certain time of the day, allowing future messages based on the recent poor performance, and adjusting the system to manage special situations more positively. As in the "rule of thumb, I believe that rules should not be too many and their number should be limited to about six.

Team
The third principle is that the trading system can not be developed collectively. Using the command sounds like an advantage, but it is not. The team will never reach consensus. Debate about the appropriateness of each rule or set to continue for a long time and most likely without substantial progress. One reason for this is that each trader has a different so-called "threshold of risk prevention." The risk, which everyone wants to personally take, and how long he is prepared to tolerate the distress situation is completely different from the other person. Everyone has a different financial situation, personality, expectations, level of patience and health. If your last transaction was actually winning, you might be overly confident and greedy. If your last transaction has brought a great loss, you might have to deal with fear and lack of confidence. Some traders are inclined to the scalp and have a large number of quick deals. Others consider the longer trends. Some are unwilling to carry positions in the other day, while others feel the discomfort of intra-day trading.

So, when designing trading systems, designs it so that it responds to you. Design it to become a better trader and to overcome your personal weaknesses. You create a useful trading system quickly and better protect themselves from the opinions, suggestions, and "wisdom" to all those who would say to you that something needs to be corrected in your system, make it more effective.

Testing
It is imperative that you have a database where you can test your trading system. This database becomes the benchmark against which each donastroyka your system can be measured. Naturally in the process of designing new ideas, and every new idea must be tested. You are trying to answer the question as to benefit a new idea or rule. It goes without saying that the framework should include examples of all types of market conditions. It would be unwise to have a test database, which moved to a variable or a trend to markets.

You want to trade a system that would work effectively in a market environment for which it was developed and survived all the others. Not every day will be profitable. A successful system - the one that supports itself, recovered slightly marginal profit loss, and accumulate large profits arising on the account. If you expect that the system will be right all the time or profitable every day, you'll never find it, and probably abstain from the many ideas that were not so bad.

Optimization
Now a few words of caution about optimization. To cite one example: I developed the trading system in 1987. When the market experienced a crash in October, which is called "Black Monday". Optimization would have to deal with any system, that it can show the entrance to the short side before the crash, go down along with the market, and then filed a message to get out of these short positions near the base. Be very careful about the optimization for the special situation.

I am not saying that such a situation as "Black Monday" should not be included in a database against which the system is tested. I mean, it would be wrong to optimize the system to surpass others in the market during the "Black Monday", because this type of market movement occurs several times in their life. I really doubt that any system developed for the "Black Monday" in 1987. still sold in 2001. in order to benefit from the collapse after the terrorist attacks on the United States. So, you should fully understand that it is optimized and the results depend on specific situations. Special situations include long-range, limited motion, GEPy at the opening and spikes.

And finally, after the development of the trading system, testing it on historical data, test it on virtual accounts for quite some time to verify its effectiveness. If it meets your expectations for some time on all types of markets, then you can start to trade on it already with real money. If your system does not work as expected, all that you have lost - it's your time, not your money. I hope that the experiences described in this article will help you in your desire to become a better trader and to develop a trading system that will work for you.

www.ensignsoftware.com