System Optimal Complexity
huangwh88:
Hello fellow traders,
I just finished 'Trading Systems: A New Approach to System Development and Portfolio Optimisation' by Tomasini and Jaekle. They mention that every system has a point at which optimal complexity is reached, meaning that further in-sample optimization will not lead to improvements in out-of-sample testing.
So my question is, at which point during development must we determine this optimal point?
I suppose it should be done prior to WFA. Lets say I have 5 parameters that can be optimized, and 10 years of historical data. I use the first 8 years for optimization, with the last 2 used for out of sample testing. After this I realize that the point of optimal complexity is reached after 3 parameters are optimized.
So with these 3 parameters, I proceed to do WFA, leaving the last 2 parameters untouched at all times. Is this the correct approach?
How you realized that optimal complexity point is reached,for that which formula or assumptions you have in your mind.it might be the rest of 2 parameters do more favorable effect on your trading action and or on your possible expectations...... for example you leave untouched "smoothing factor" that can save you false signaling/order execution,do you think it help you or you like let it untouched ? Hello fellow traders,
I just finished 'Trading Systems: A New Approach to System Development and Portfolio Optimisation' by Tomasini and Jaekle. They mention that every system has a point at which optimal complexity is reached, meaning that further in-sample optimization will not lead to improvements in out-of-sample testing.
So my question is, at which point during development must we determine this optimal point?
I suppose it should be done prior to WFA. Lets say I have 5 parameters that can be optimized, and 10 years of historical data. I use the first 8 years for optimization, with the last 2 used for out of sample testing. After this I realize that the point of optimal complexity is reached after 3 parameters are optimized.
So with these 3 parameters, I proceed to do WFA, leaving the last 2 parameters untouched at all times. Is this the correct approach?
huangwh88:
Hello fellow traders,
I just finished 'Trading Systems: A New Approach to System Development and Portfolio Optimisation' by Tomasini and Jaekle. They mention that every system has a point at which optimal complexity is reached, meaning that further in-sample optimization will not lead to improvements in out-of-sample testing.
So my question is, at which point during development must we determine this optimal point?
I suppose it should be done prior to WFA. Lets say I have 5 parameters that can be optimized, and 10 years of historical data. I use the first 8 years for optimization, with the last 2 used for out of sample testing. After this I realize that the point of optimal complexity is reached after 3 parameters are optimized.
So with these 3 parameters, I proceed to do WFA, leaving the last 2 parameters untouched at all times. Is this the correct approach?
Hello fellow traders,
I just finished 'Trading Systems: A New Approach to System Development and Portfolio Optimisation' by Tomasini and Jaekle. They mention that every system has a point at which optimal complexity is reached, meaning that further in-sample optimization will not lead to improvements in out-of-sample testing.
So my question is, at which point during development must we determine this optimal point?
I suppose it should be done prior to WFA. Lets say I have 5 parameters that can be optimized, and 10 years of historical data. I use the first 8 years for optimization, with the last 2 used for out of sample testing. After this I realize that the point of optimal complexity is reached after 3 parameters are optimized.
So with these 3 parameters, I proceed to do WFA, leaving the last 2 parameters untouched at all times. Is this the correct approach?
You can not determine that point
It is purely speculative that "optimal complexity is reached" and there is no criteria that could help you in that
In real-time processing of complex systems where you have no physical laws between input and output (only deterministic and stochastic realations) you have an amount of parameters where you have an optimum of accuracy and speed. There you know it from your programm or after some tests where the optimum is. Here in FX or other financial market you have no programm or matrices that describe the system, you only have some tools which work good because many traders use them. Combine them and you will get the experience you need to decide whether you need one more filter or kind of pattern or whatever. Too many and your head is overstrained, too less and you'll miss information you need to trade. To use indicators for signals is handcraft, real trading is art.
krelian99:
In real-time processing of complex systems where you have no physical laws between input and output (only deterministic and stochastic realations) you have an amount of parameters where you have an optimum of accuracy and speed. There you know it from your programm or after some tests where the optimum is. Here in FX or other financial market you have no programm or matrices that describe the system, you only have some tools which work good because many traders use them. Combine them and you will get the experience you need to decide whether you need one more filter or kind of pattern or whatever. Too many and your head is overstrained, too less and you'll miss information you need to trade. To use indicators for signals is handcraft, real trading is art.
And there is always the danger of over-optimizing the whole system too
In real-time processing of complex systems where you have no physical laws between input and output (only deterministic and stochastic realations) you have an amount of parameters where you have an optimum of accuracy and speed. There you know it from your programm or after some tests where the optimum is. Here in FX or other financial market you have no programm or matrices that describe the system, you only have some tools which work good because many traders use them. Combine them and you will get the experience you need to decide whether you need one more filter or kind of pattern or whatever. Too many and your head is overstrained, too less and you'll miss information you need to trade. To use indicators for signals is handcraft, real trading is art.
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
I just finished 'Trading Systems: A New Approach to System Development and Portfolio Optimisation' by Tomasini and Jaekle. They mention that every system has a point at which optimal complexity is reached, meaning that further in-sample optimization will not lead to improvements in out-of-sample testing.
So my question is, at which point during development must we determine this optimal point?
I suppose it should be done prior to WFA. Lets say I have 5 parameters that can be optimized, and 10 years of historical data. I use the first 8 years for optimization, with the last 2 used for out of sample testing. After this I realize that the point of optimal complexity is reached after 3 parameters are optimized.
So with these 3 parameters, I proceed to do WFA, leaving the last 2 parameters untouched at all times. Is this the correct approach?