In previous work I have done- I have found it very useful in either system or indicator optimization to have 'in-sample' and 'out-of-sample' dates for optimization.
I found that specifying 2000-2005 as in-sample and 2006 as out-sample was much more likely to result in over-optimization rather than using the entire date range for optimization and choosing random months.
This of course does not apply to all optimizations (sometimes you want recent dates optimized fully etc).
Let me see if I can explain. What you want is for a given date range that you are optimizing for (daily,weekly,monthly) is to have randomly within that date range- in-sample and out-of-sample dates.
The random is important. After optimization on in-sample- you can confirm on out-of-sample.
What I did in my optimizer I designed was have an check button next to a date range and then a drop down of % dates to include.
So for example the user chooses 1/1/2000 to 1/1/2011 for optimization next to it is option to check 'random months %' . He checks it then chooses 70%.
The optimizer then takes the total number of months- and randomly throws out 30%.
Then internally during the optimization run I just closed all positions at last day of month if the next month was 'out-of-sample'.
Now the problem is after optimizing you want to see the out-of-sample dates in your date range.
So what I did is have a function elsewhere in program to specify in-sample and out-of-sample dates and store them.
So you would have a program to generate a list of in-sample and out-of-sample dates for months/days etc.
The user would choose full date range- percentage of out-of-sample etc and two files would be made with dates to include.
Then in any optimization procedure the user would specify 'use in-sample monthly dates' or 'in-sample daily dates' for optimization.
Then when simulating a trading system- you could confirm after all your optimizations on 'out-of-sample' dates.
Anyway- there are many different ways to do this- I hope this gives you some ideas. Thanks. Marc
Trading financial instruments, including foreign exchange on margin, carries a high level of risk and is not suitable for all investors. The high degree of leverage can work against you as well as for you. Before deciding to invest in financial instruments or foreign exchange you should carefully consider your investment objectives, level of experience, and risk appetite. The possibility exists that you could sustain a loss of some or all of your initial investment and therefore you should not invest money that you cannot afford to lose. You should be aware of all the risks associated with trading and seek advice from an independent financial advisor if you have any doubts.
0
