(Posted this in the wrong section, sorry)
So I have been programming Expert Advisors for a few months now and have had some success. My latest program yields some interesting results. I just wanted to get some input on my results from you guys - let me know what you think:
Market keeps changing weekly monthly yearly, you have been successful so far does not mean you will be continuously successful in the next couple of weeks, months, years.
I think you need to come back 6 years later once you made 2.5 mill then create few thread with such name
+1 Markaria400
See that little box that says “modelling quality n/a”
Means your results are null and void. Don’t give up the day job bro!
bobbillbrowne I don’t believe you get a number for modelling quality if you use anything higher than 1M do you?
on a side note have you done a comparison between using 1M data and tick data?
Looks like it’s time to go update your data bro.
haha I’ve just been reading this. I will have a go and retest.
Hey Guys,
I have no skills in writing EA’s but I reckon many EA developers have been going about developing an EA the wrong way because nearly every EA I’ve read about have at some time stopped working or crashed because the market conditions had changed. I reckon the answer in developing the ideal EA has been staring us in the face. We should be developing an EA to suit the different market conditions i.e. analyse and evaluate all possible market conditions - lets say we find 20 different market conditions that are common and occurring on a weekly basis. We write an EA for each of the 20 different market conditions and then write a trigger EA to control all 20. The job of the trigger EA is simple, analyze the current market condition and if the results satisfy one of the 20 EA’s, execute that EA and keep running it until the trigger EA detects a change in the market condition. When that happens turn off the first EA, and run the EA that is written for the new market condition. So basically the trigger EA is turning EA’s on and off according to the market condition. As a safeguard we include a condition in the trigger EA that turns off all trading activity when it detects a market condition that doesn’t satisfy our list of twenty EAs
Taking this scenario a bit further, we can make the trigger EA keep checking the market conditions continuously and take the appropriate action when necessary using our list of 20 market conditions. Alternatively we could make the market conditions list a bit bigger.
That’s my 2 cents worth
Well I usually keep my strategies separate from each other. 1 EA for every strategy/currency. I like to be able to quickly test new ideas so i keep a template with a nice flow, and so if i started mixing everything up i would have loads of input variables and i wouldn’t be able to keep track of it.
Hi Mark,
At least it sounds like you might have something interesting going there. Have to tried forward testing your EA yet? I certainly would be interested to see how you get on… I like seeing people “make it big”. Good luck, and keep us posted!
I’ve started to work on getting 99.99% modelling quality. I thought for starters I should go from using 15M timeframe to 1M by adjusting the EA, and the results don’t really differ. This is not a Scalping system; positions are open for up to 2 days and focus on larger price movements.
But when I think of what is actually getting fed into the EA via the OPEN HIGH LOW CLOSE values, I don’t think getting into actual tick resolution will help much. Suppose the market moves against me with a BUY position is open, if the LOW of the minute bar hits my stop loss – it closes doesn’t matter if the high was first (since that’s quite far from the open), end of that.
I understand that 99.99% modelling quality is important for some systems, but I’m not convinced for this system. But I will still give it a go.
I have a number of EAs running at the moment, but its a slow process.
so i bet you are now half way to millionaire
Well ive taken bobbillbrowne and moved to tick data to double check the results. I did this by transposing the EA into cTraders cAlgo suite (As you can test using tick data without having to play around with thrid party programs). The results are almost the same, although I didn’t use the same money management or compounding lot sizes.
I guess the best thing to do now is to leave it running and start a trade journal.