Multiple timeframes and backtesting. Workaround
I have an EA for Period 5 min but in the code I also use daily data like the daily 8 and 21 SMA. I want to backtest my EA but since MT4 cannot backtest multitimeframe I need a workaround for this:
Yes it can.
hmm interesting. I'm new to programming but I just started using Molanis and on their website they write:
"Multi timeframe EAs CANNOT be backtested in MT4. It is not possible. This ea will work fine but backtesting results are misleading. MT4 can backtest one timeframe at a time. The only way to backtest multi timeframe EAs is using MetaTrader 5" http://www.molanis.com/products/molanis-strategy-builder/expert-advisors-for-mt4
Why would they write this if its wrong?
I guess you know what youre talking about, but what is the trick to make it work? Do I need 99% modeling quality? I use 90
hmm interesting. I'm new to programming but I just started using Molanis and on their website they write:
"Multi timeframe EAs CANNOT be backtested in MT4. It is not possible. This ea will work fine but backtesting results are misleading. MT4 can backtest one timeframe at a time. The only way to backtest multi timeframe EAs is using MetaTrader 5" http://www.molanis.com/products/molanis-strategy-builder/expert-advisors-for-mt4
1. Why would they write this if its wrong?
I guess you know what youre talking about, but what is the trick to make it work? Do I need 99% modeling quality? I use 90
1. you will have to ask them: first read this: Testing Features and Limits in MetaTrader 4
You only get 99% modelling quality using tick data and a 3rd party paid for application, and then 99 is just the number written to a file, it doesn't mean anything.
Perhaps they are talking about this:
"Zero bar of another timeframe for the same symbol under test is modeled approximately
Open = correct Open, Close = correct Close, Low = min (Open,Close), High = max (Open,Close), Volume = final Volume (false)"
it may be a problem if you use Indicators that re-paint . . . which you seem to be doing in your post above.
1. you will have to ask them: first read this: Testing Features and Limits in MetaTrader 4
You only get 99% modelling quality using tick data and a 3rd party paid for application, and then 99 is just the number written to a file, it doesn't mean anything.
Thank you very much. I've done som simple testing now and it works.
Hello coders,
I have somehow the same issue with a little difference so I decided to comment here:
I wonder if I could use multi-time frame data for the same Symbol when I'm backtesting my EA. Not just generated OHLC, but its fully tick data!
I've purchased tick data(99%) but I don't know how can I test my EA on 2 different time frames at the same time since I'm Newbie...
For example I want to open order whenever 4H and 1H candles both hit their Moving Averages.
Any advice is appreciated,
Hossein
- 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 have an EA for Period 5 min but in the code I also use daily data like the daily 8 and 21 SMA. I want to backtest my EA but since MT4 cannot backtest multitimeframe I need a workaround for this:
I have this daily SMA filter:
iMA(Symbol(),PERIOD_D1,8,0,MODE_SMA,PRICE_CLOSE,0) < iMA(Symbol(),PERIOD_D1,21,0,MODE_SMA,PRICE_CLOSE,0)
I need to rewrite so the daily 8 and 21 SMA is calculated from the 5 min chart. I've tried to put in 2304 and 6048 instead of but it dosen't seem to work:
iMA(Symbol(),0,2304,0,MODE_SMA,PRICE_CLOSE,0) < iMA(Symbol(),0,6048,0,MODE_SMA,PRICE_CLOSE,0
I also use this formula and need to rewrite so it works with 5 min backtest:
iHigh(Symbol(),PERIOD_D1,0)-iLow(Symbol(),PERIOD_D1,0)
Same with this, find daily ATR:
iATR(Symbol(),PERIOD_D1,10,0)
I would really appreciate if somebody can help or give me som hints.