Yoghurt systems and canned systems or The relationship between trading tactics and the reliability of historical test results
I would like to propose a discussion of the relationship between trading tactics and the reliability of historical testing results.
When we create a trading system, we rely on some regularities that depend on the input variables. We often optimize the system according to these parameters at the testing stage and rejoice when we get good results. And then it turns out that the system starts to fail after a week. Some systems only start to fail after a few months. I suggest and discuss the factors that affect the "shelf life" of systems. Why some systems are yoghurt, others canned tuna, and is it possible to create a perpetuum mobile or at least come close to it?
I have proposed this topic, in a slightly more general way, for discussion... :)) Attempt number 2. Or let's talk about earning, but without being specific ... :)) But the result will not be. 90% will not understand what we are talking about at all. :) But I hope that I was wrong and there will be some sense in your theme.
then the comb on the graph == 100% yoghurt.
An EA should not have any parameters that need to be optimized (picked up on history). IHMO optimisation on history is cheating itself. The only thing that I think is acceptable is if in the whole range of changes from - nocon to + nocon. The Expert Advisor remains profitable (all runs) then it is worth choosing a parameter from the area that ensures a stable maximum profit.
The patterns that exist in the market cannot be influenced by your "input variables"
Of course it does. We either see a pattern or we don't. If we identify it correctly, we have profits. The input variables do not affect the general patterns of the market, which is clear, but what we see as signals.
Well, it goes on to say: the shelf life of the system, i.e. the rate of obsolescence of the parameters.
I would like to propose for discussion the issue of the correlation between trading tactics and the reliability of the test results on history.
When we create a trading system, we rely on some regularities that depend on the input variables. We often optimize the system according to these parameters at the testing stage and rejoice when we get good results. And then it turns out that the system starts to fail after a week. Some systems only start to fail after a few months. I suggest and discuss the factors that affect the "shelf life" of systems. Why are some systems yoghurt, others canned tuna, and is it possible to create a perpetuum mobile or at least come close to it?
Nicely phrased "we rely on some kind of regularities that depend on the input variables" This is the crux of the whole problem!
I think the longevity of a system depends on what regularities it is built on. The market is volatile and the patterns that worked before may not work now. There are different kinds of patterns in the market. Some last a long time, others do not reveal themselves much, some are not regularities at all, but wishful thinking. Therefore, to create a reliable system, you have to immediately eliminate signal systems based on short patterns. You have to look for long patterns and catch signals from them.
How's that? When you write trade conditions that depend on the indicator parameters, don't you process "repeating situations"? And if you change the indicator parameters, what then? Then you will consider somewhat different patterns. In short, it depends on the parameters which patterns you will be looking at.
If the optimisation graph is combed it's a sure sign that MTS depends on parameters; parameters need tuning into resonance,
then the comb on the graph == 100% yoghurt.
I agree, the "run" patterns after the signal (StopLoss, TakeProfit) are probably the most "perishable" patterns. "Row" systems tend to go out on these orders and are therefore yoghurt.
You think there are no perishable patterns on D1? :) I'm sure they exist, they just manifest themselves in proportion to the TF. And leaving to such scales greatly limits a trader, don't you agree?)
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
I would like to propose a discussion of the relationship between trading tactics and the reliability of historical testing results.
When we create a trading system, we rely on some regularities that depend on the input variables. We often optimize the system according to these parameters at the testing stage and rejoice when we get good results. And then it turns out that the system starts to fail after a week. Some systems only start to fail after a few months. I also suggest discussing factors that affect the "shelf life" of systems. Why some systems are yoghurt, others canned tuna, and is it possible to create a perpetuum mobile or at least come close to it?