Advertisement
Looking for Metatrader Backtesting Guru ( MT4)
Jan 30, 2015 at 18:24
726流覽
8 Replies
forex_trader_149646
會員從Sep 09, 2013開始
406帖子
Jan 30, 2015 at 18:24
I am building systems located here - http://www.myfxbook.com/members/RedRhinolab and I have a large list of systems that I'll be developing in 2015. I would like to meet someone who is a backtesting guru to optimize these systems. Quality backtests take a lot of time and resources. You will be rewarded by a free license of each system.
forex_trader_149646
會員從Sep 09, 2013開始
406帖子
Jan 30, 2015 at 18:32
Optimization through backtesting can "Open The Window" to a completely new system than the intended original design. For example, a system designed to run on a Daily Time frame can be ran on a H1 timeframe with different setting and produce far better results. The same goes for a system that was designed for a H1 TF can be implemented on a M5 TF with tighter SL,TP,TS to produce extraordinary results. Or you can optimize the original design to find better settings that actually change the system completely.
forex_trader_149646
會員從Sep 09, 2013開始
406帖子
Jan 30, 2015 at 18:40
Qualification:
Tick data - It's not necessary for 99% Modeling Quality unless optimizing trailing stops, but the tick data from a quality source "http://www.tickstory.com" or equivalent should be used to create M1 Bars that are then "Period Converter" to all the standard timeframes. By doing so, data holes don't exist and the backtester has the ability to work correctly. If there are better methods today, please discuss them as well.
Tick data - It's not necessary for 99% Modeling Quality unless optimizing trailing stops, but the tick data from a quality source "http://www.tickstory.com" or equivalent should be used to create M1 Bars that are then "Period Converter" to all the standard timeframes. By doing so, data holes don't exist and the backtester has the ability to work correctly. If there are better methods today, please discuss them as well.
會員從Nov 21, 2011開始
1601帖子
Jan 30, 2015 at 20:14
There is noting better to test real demo account over months.
It takes time to test a strategy.
It takes time to test a strategy.
forex_trader_149646
會員從Sep 09, 2013開始
406帖子
Jan 31, 2015 at 06:24
Yes I agree, Live tests are better but the task is to unlock the potential of a system and through backtesting with optimization we can satisfy this task. It is truly amazing what we can uncover when backtesting, with a walk forward testing approach.
http://en.wikipedia.org/wiki/Walk_forward_optimization
http://en.wikipedia.org/wiki/Walk_forward_optimization
會員從Nov 21, 2011開始
1601帖子
Jan 31, 2015 at 10:27
In fact you simply need to know if your system was designed for trendy market or sideway market.
When you find or developp your own Indicator that can indicate such information, you allow your EA to trade or not.
If you think your strategy is suitable for any kind of market conditions, then you let it tun all time but you have different parameters to adjust.
When you find or developp your own Indicator that can indicate such information, you allow your EA to trade or not.
If you think your strategy is suitable for any kind of market conditions, then you let it tun all time but you have different parameters to adjust.
會員從Aug 30, 2012開始
64帖子
Jan 31, 2015 at 14:07
My System, can not be back tested, for a lot of reasons but one simple example is because its reading Internet News feeds like Reuters ....
LIKE MY TRADE ALERT POST! TRY IT FREE! NO SIGN UP! 100% FREE - w.w.w.ForexAlertSystem.c.o.m "CLICK ON BIG GREEN TRIAL BUTTON"
會員從Jan 22, 2014開始
1帖子
Feb 09, 2015 at 07:38
I'm interested in backtesting your systems. I use a customized backtesting engine built on a C++ framework that utilizes walk forward analysis. I also have access to high quality historical data.
Let me know if you're interested.
Let me know if you're interested.
*商業用途和垃圾郵件將不被容忍，並可能導致帳戶終止。
提示：發佈圖片/YouTube網址會自動嵌入到您的帖子中！
提示：鍵入@符號，自動完成參與此討論的用戶名。
