- Search Metals Mine
- 24 Results
- James.Zhang replied Sep 21, 2016
From the open positions, we could know that the EA thinks that it is time to short US Dollar, to long others including indexes, Copper, Gold, Silver, EUR, JPY, etc. Now there are 9 JPY long positions, 3 EUR long positions and several indexes, ...
- James.Zhang replied Sep 20, 2016
Yes, my tester is only for my own use. I use the data from FXCM's API, though there are some data being not correct, some duplicated, they are still good as a whole.
- James.Zhang replied Sep 19, 2016
Oh... Sorry for that. Can not do the conversion between mql4 and C++.
- James.Zhang replied Sep 17, 2016
I met the same issue as you before. The only way is to give up MT4, developing my own backtesting software, supporting multiple pairs, with excellent execution time.
- James.Zhang replied Sep 16, 2016
Till now, 66 trades have been closed. 24 profit trades, 42 loss trades, Win Rate is 36.36%, which is close the my backtesting result 38.2%. The interesting thing is that almost all the profit is in open trades, not the closed one. That's how I apply ...
- James.Zhang replied Sep 14, 2016
5~8 pairs is best for normal people. You will be lost if too much.
- James.Zhang replied Sep 14, 2016
For this EA, so far, so good. It opened an SPX500 long position today. That's interesting. I wouldn't do that if I do manually. Below is the equity curve till now. image
- James.Zhang replied Sep 14, 2016
Hoping we have chance to share our ideas.
- James.Zhang replied Sep 14, 2016
One more short for GBP/USD and long for EUR/JPY supporting USD/JPY long position. image
- James.Zhang replied Sep 13, 2016
Gook luck, my friend. Nice meeting you.
- James.Zhang replied Sep 13, 2016
My EA is a standalone program running on Windows/Linux. I think it can not call MQL based indicators/exe, which can only be called from MT4. Of course I can share one of my backtesting result. file . But it is in my own format. I just generate the ...
- James.Zhang replied Sep 13, 2016
Vee, You are totally right. Till now, you are the first one understanding what I am doing in the past 4 months. Thank you. Actually, my strategy is pretty simple, like what I've been using for the past several years. It's something like MACD golden ...
- James.Zhang replied Sep 11, 2016
Oh.. I have some other accounts running my EA for more than 2 months. You are right. I need to be more focusing. It is a long way for me, I just started. Yes, every EA has different technical setup, mainly different SL range and different enter/exit ...
- James.Zhang replied Sep 11, 2016
Yes, curve fitting is a big topic, which is what I try to prevent from the beginning of EA design. I only use historic data to generate two parameters of my EA, one it the maximum SL, the other is the exit timeframe. I treat the SL range and ...
- James.Zhang replied Sep 11, 2016
Yes, my trade profit/loss percentage is about 38%, so lots of loss trades. My profit comes from "cut loss, hold profit as long as possible", you know, from backtesting result, the average holding days of loss trade is about 1.7 days, the average ...
- James.Zhang replied Sep 11, 2016
Hi Vee, I think it is not hard to reprogram your EA with broker's API based on your current MT4 EA. Maybe 40% of the code can be reused. I finished my first version from scratch within 1.5 months. I think if we want to run EA like what we do, API is ...
- James.Zhang replied Sep 11, 2016
Thanks Vee, for question 1 and 2, it is because I am developing my EA actually from scratch based on brokers' API. I can do whatever I want, I programmed my own backtest program which can support any number of EA's backtest for the same account. ...
- James.Zhang replied Sep 11, 2016
Interesting videos. From my opinion, the bigger the timeframe, the more consistent the outcome.
- James.Zhang replied Sep 11, 2016
Good to know that. Maybe we can discuss something based on this, like how to reduce the MDD with multiple timeframes/multiple assets, how to determine the money management, etc.
- James.Zhang replied Sep 11, 2016
Yes, Curve Fitting is my most concern. I've run this EA on some live accounts for about 2 months, although no margin call, its return is either not as good as that in backtesting. Never try monte carlo simulation before, is it really useful for us? ...