Welcome to Forex-TSD!, one of the largest Forex forums worldwide, where you will be able to find the most complete and reliable Forex information imaginable.
From the list below, select the forum that you want to visit and register to post, as many times you want. It’s absolutely free. Click here for registering on Forex-TSD.
Exclusive Forum
The Exclusive Forum is the only paid section. Once you subscribe, you will get free access to real cutting-edge Trading Systems (automated and not), Indicators, Signals, Articles, etc., that will help and guide you, in ways that you could only imagine, with your Forex trading.
Elite Section
Get access to private discussions, specialized support, indicators and trading systems reported every week.
Advanced Elite Section
For professional traders, trading system developers and any other member who may need to use and/or convert, the most cutting-edge exclusive indicators and trading systems for MT4 and MT5.
There is a bug which make the backtests useless; in two words, the EA knows the future when it uses the daily values High, Low and Close : those values are those of the closed daily bar. This means that at 8 AM it already knows what will happen. But there is no meaning to use the previous bar.
Another problem is to define a broker independent day start.
Another bug is to assign a zero value if the range is null (ie the first tick of the day); it should be 50. Anyway, the values aren't very meaningfull at the begining of the day, when the range is not yet wide. Maybe a trailing time range would be better.
Also, it seems to me that the values should be normalized and maybe wheigted by the number of pairs giving an information.
I am currently working on all that and will keep you informed.
Nevertheless, forward testing should be interesting...
Backtest is useless??? because they know the future from daily values High, Low and Close ??? . Are you trying to say that this ea was only good in backtesting???
The backtesting for EU from june 2004 to jan 2005 shown a very good smooth climbing curve. The bigest drawdowns was about 60% from your ac ac shown in circle.
The only doubt to me was why some of the situation this backtest seems like in idle mode(no trade open) sometime for 1 day even though from the candle it shown quit heavy movement. You can see it from post # 403 to 404 below graf.
Backtest is useless??? because they know the future from daily values High, Low and Close ??? . Are you trying to say that this ea was only good in backtesting???
Not at all, I just said that backtests may seem good because of this bug.
Up to know, I do not know if without this bug it may be good in live or in BT.
But, as there aren't so much parameters allowing curve fitting, forwards tests should show the real profitability.
Not at all, I just said that backtests may seem good because of this bug.
Up to know, I do not know if without this bug it may be good in live or in BT.
But, as there aren't so much parameters allowing curve fitting, forwards tests should show the real profitability.
To confirm whither there was no trading(idle) at certain period is due to no signal or it was actually a bug, i think the best way is can somebody design an indicator that able to draw a line of signal bar meter for both currency being traded.
To confirm whither there was no trading(idle) at certain period is due to no signal or it was actually a bug, i think the best way is can somebody design an indicator that able to draw a line of signal bar meter for both currency being traded.
For 2 currency being traded, there need checking about 8 other currencies.
which is not possible to have a BT for this system.
What if we stored the difference of the indicator (ex. 7-2=5) say of EURUSD everytime it changes then at the end of the day we generate an average and show it as a guide for the next day. That way we have a trend indicator.
What if we stored the difference of the indicator (ex. 7-2=5) say of EURUSD everytime it changes then at the end of the day we generate an average and show it as a guide for the next day. That way we have a trend indicator.
No, the !xMeter react quickly with the market and even an average couldn't be accurate for the next day. I think ...
When I read few first posts about backtest I knew you would get this problem - when backtesting using other currencies it looks into the future.
And yes, I know how to avoid this problem. But it will require for me (you?) to re-write the EA completely and use M1 data instead of D1. So it will only look 1 minute ahead. Or you will make it look one minute back so it won't hurt as badly as H1 or D1.
There is no other way as it seems.
What you need is making it calculate D1 data using H1 data or even M1 data. I already found a way to find daily high/low/close using M1 data. But when I asked about the indicators that can calculate data for other periods I only got criticized. I guess that is why successful traders rarely share their ideas - they are not understood.
I'll try to help if you can make your need clear. Oh and its useless to try and PM me - my inbox is almost full and I don't know what to delete there. If you want to contact me, use ICQ instead.
When I read few first posts about backtest I knew you would get this problem - when backtesting using other currencies it looks into the future.
And yes, I know how to avoid this problem. But it will require for me (you?) to re-write the EA completely and use M1 data instead of D1. So it will only look 1 minute ahead. Or you will make it look one minute back so it won't hurt as badly as H1 or D1.
There is no other way as it seems.
What you need is making it calculate D1 data using H1 data or even M1 data. I already found a way to find daily high/low/close using M1 data.
The solution is already here: indicator values from other timeperiod
Other point: the xMeter indic shows nothing but the stochastic on a variable period (from the start of the day), but I am sot sure that this is very interesting, as the value is not very significant at the day's begining.
It should be better (and easier) to use a fixed period of time for the stochastic (for sample, 24 h).
The solution is already here: indicator values from other timeperiod
Other point: the xMeter indic shows nothing but the stochastic on a variable period (from the start of the day), but I am sot sure that this is very interesting, as the value is not very significant at the day's begining.
It should be better (and easier) to use a fixed period of time for the stochastic (for sample, 24 h).
That would really help.
Oh by the way, since I didn't look into the code, but maybe you'll find this useful:
. Are you trying to say that this ea was only good in backtesting???
Linear Mode
