Systems traders always test their systems thoroughly before they use them and if they make any changes.
When you are putting the system rules and parameters into your system, you have to test it to check if it works as it should. You make sure that it does all the basic things correctly, such as entering and exiting trades.
This is a simple and straightforward task and you can use any price information because you are only testing the mechanical way your system works. These tests do not have any value beyond that – they give you no idea how the system will operate in a live trading situation.
When you know you have put your system together and it works, you can then get down to finding out its capabilities by back testing with historical prices.
Your tests will need to simulate live conditions as closely as possible. Your software is the key element in this as it will need to process every single day in your testing period as if it was a real live day, before moving on the next day. This is the only way you can re-live all the situations that would have arisen.
Results from the back testing give you vital information on the capabilities of your system and allow you to see how it would have performed in different situations. You will be able to obtain answers to important questions that may help you to improve your system, such as: - Did it cut losing trades – did it run winning trades? What are its strengths and weaknesses?
Back testing is an essential part of the evaluation process system traders carry out to assess the suitability of any system that interests them.
The tests are usually conducted with historical prices for the previous 5 to 10 years, which are obtainable from data providers.