WHICH METHOD REFLECTS REAL TRADING BETTER?
WHICH MODEL. TICK, CONTROL, OR OPEN PRICES REFLECTS MORE CLOSELY THE RESULTS OBTAINED IN REAL TRADING?
- Green Bars on Report
- EA Model. Help Me Please!
- Experts: 1H Bolinger Bands
Paul,
Typing in Caps is considered as shouting at people in cyberspace and therefore you might find some people will avoid helping you.
Personally, the Tick method I find to be the better results as it endevours to reflect the tick based model of a data feed. It's also the longest test method due to the number crunching. You'll also need a good amount of 1 minute data to obtain the best results. You're able to source 1 minute data from Alpari. Do a search of this forum for the URL.
As always, backtesting isn't a reflection of the future and there's a few caveats and gotchas in relation to how an EA is written in how it behaves in a backtest. Again, search this forum and read the MQ FAQs for more info.
Cheers
Martin
Typing in Caps is considered as shouting at people in cyberspace and therefore you might find some people will avoid helping you.
Personally, the Tick method I find to be the better results as it endevours to reflect the tick based model of a data feed. It's also the longest test method due to the number crunching. You'll also need a good amount of 1 minute data to obtain the best results. You're able to source 1 minute data from Alpari. Do a search of this forum for the URL.
As always, backtesting isn't a reflection of the future and there's a few caveats and gotchas in relation to how an EA is written in how it behaves in a backtest. Again, search this forum and read the MQ FAQs for more info.
Cheers
Martin
Martin be careful with Alpari 1min. They added some spread or something to quotes and fractal testing brings noise and corrupts results.
Paul. Testing models are supposed to be gradually accurate in real market approximation.
OPEN PRICES - should be less precise.
CONTROL POINTS with lower frame - better
FRACTALS with less time frame avail. - should be the most precise - almost like real market ticks.
For my experts it is true, but imagine that your experts work on open prices only... :)
So I think it depends on few things, like:
- prices used in expert;
- quality of historical quotes values
- expert philosophy
There is also an article about testing models on this site. Perhaps it would be useful.
"Strategy Tester: Modes of Modeling during Testing"
Paul. Testing models are supposed to be gradually accurate in real market approximation.
OPEN PRICES - should be less precise.
CONTROL POINTS with lower frame - better
FRACTALS with less time frame avail. - should be the most precise - almost like real market ticks.
For my experts it is true, but imagine that your experts work on open prices only... :)
So I think it depends on few things, like:
- prices used in expert;
- quality of historical quotes values
- expert philosophy
There is also an article about testing models on this site. Perhaps it would be useful.
"Strategy Tester: Modes of Modeling during Testing"
Paul,
Typing in Caps is considered as shouting at people in cyberspace and therefore you might find some people will avoid helping you.
Personally, the Tick method I find to be the better results as it endevours to reflect the tick based model of a data feed. It's also the longest test method due to the number crunching. You'll also need a good amount of 1 minute data to obtain the best results. You're able to source 1 minute data from Alpari. Do a search of this forum for the URL.
As always, backtesting isn't a reflection of the future and there's a few caveats and gotchas in relation to how an EA is written in how it behaves in a backtest. Again, search this forum and read the MQ FAQs for more info.
Cheers
Martin
Typing in Caps is considered as shouting at people in cyberspace and therefore you might find some people will avoid helping you.
Personally, the Tick method I find to be the better results as it endevours to reflect the tick based model of a data feed. It's also the longest test method due to the number crunching. You'll also need a good amount of 1 minute data to obtain the best results. You're able to source 1 minute data from Alpari. Do a search of this forum for the URL.
As always, backtesting isn't a reflection of the future and there's a few caveats and gotchas in relation to how an EA is written in how it behaves in a backtest. Again, search this forum and read the MQ FAQs for more info.
Cheers
Martin
Thanks for the advice on the caps
regarding the essence of the matter, I understand the diference, but from the experience standpoint, how do they , each method, compare to real results?
None of them will be wholly accurate in regard to real time trading. However, if you have enough 1min data for the entire timeframe you are testing, you should be able to get reasonable results. YOU MUST download and import A LOT of data on your own to get accurate testing, you simply will not have enough 1 minute data for accurate testing otherwise if you are looking for reasonable backtesting accuracy.
Alparis databank is excellent: http://www.alpari-idc.com/en/dc/databank.php
That being said, the longer timeframe you trade on, with the wider stops/targets will also allow for more accurate test results, because these wider spreads between your targets, etc. won't be as likely to be affected by small price movements that aren't tested accurately. Trying to backtest a real fine scalper will prove very difficult because you will not have all the price movements.
Alparis databank is excellent: http://www.alpari-idc.com/en/dc/databank.php
That being said, the longer timeframe you trade on, with the wider stops/targets will also allow for more accurate test results, because these wider spreads between your targets, etc. won't be as likely to be affected by small price movements that aren't tested accurately. Trying to backtest a real fine scalper will prove very difficult because you will not have all the price movements.
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