OzFx System :) 100-800 Pips Per Trade - page 5
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
Azmel,
You can just use the backtester to simulate the trade, and watch it visually. No need to wait days to figure out if your code is working as expected. That's how I troubleshoot, and it works well especially for slow traders.
Hello OzFx,
I just want to recall the fxbs remarks about the Stoch crossing the 0 line of AC. The problem I see is that the 0 crossing point depends on the range, the zoom and the chart shift of your display since moving the chart to another bar moves the 0 crossing of the Stoch up or down!!! Try it you will see. It does not move much when AC is quite in the same range but when big bumps happens, the entry signal can be way off from one person to another.
I don't know if you were aware and understand the problem and I just want to bring this up to make sure there is no mistakes here.... If there are EAs working out there, then they must take into account the range of the AC for a certain amount of bars (how many?) and then adjust the Stoch relative position accordingly and then compare it to 0 of the AC. Otherwise it is not working pretty well.
Have a look at the charts. They are the exact same charts but one has more bars than the other on the display. See the Stoch crossing moved a little to the left on the second one and the relative position is different...
Maybe this is part of your setup but then we need to know how many bars you have on your screen !!
Hello OzFx,
I just want to recall the fxbs remarks about the Stoch crossing the 0 line of AC. The problem I see is that the 0 crossing point depends on the range, the zoom and the chart shift of your display since moving the chart to another bar moves the 0 crossing of the Stoch up or down!!! Try it you will see. It does not move much when AC is quite in the same range but when big bumps happens, the entry signal can be way off from one person to another.
I don't know if you were aware and understand the problem and I just want to bring this up to make sure there is no mistakes here.... If there are EAs working out there, then they must take into account the range of the AC for a certain amount of bars (how many?) and then adjust the Stoch relative position accordingly and then compare it to 0 of the AC. Otherwise it is not working pretty well.
Have a look at the charts. They are the exact same charts but one has more bars than the other on the display. See the Stoch crossing moved a little to the left on the second one and the relative position is different...
Maybe this is part of your setup but then we need to know how many bars you have on your screen !!
Hi pegasus200,
You have a valid point. That is a challenge when you do visual backtest. However when you forward test it becomes much more clear. Also read through my AES post on my blog where we enter trades when Stot crossed below 0 line.
Advance Entry Signals (AES)
Hope it is not against the rules to post links here.
Cheers,
OzFx
Stochastic crossing AC
It is quite difficult to code an EA that knows when the Stochastic line crosses the AC zero line, unfortunately. Visually it is very easy to see, but ask a computer to see it, and it can't.
For a computer, the Stochastic value ranges from 0 to 100 as fxbs has pointed out. Therefore for the EA, I take the average value, which is 50 as the zero line. For optimisation, I also put the option to adjust the upper and lower values just to see what the effects of those will be on the trades.
Talking about the EA, it is almost ready. Last night, I tested it against a smaller timeframe to see what happens when a new signal appears. As expected, the fifth trade closes and a new set of five trades re-appears.
I will post the EA today for everyone to test. However, DO NOT GO LIVE with this EA yet as I have not incorporate the 200SMA filter.
It is quite difficult to code an EA that knows when the Stochastic line crosses the AC zero line, unfortunately. Visually it is very easy to see, but ask a computer to see it, and it can't.
For a computer, the Stochastic value ranges from 0 to 100 as fxbs has pointed out. Therefore for the EA, I take the average value, which is 50 as the zero line. For optimisation, I also put the option to adjust the upper and lower values just to see what the effects of those will be on the trades.
Talking about the EA, it is almost ready. Last night, I tested it against a smaller timeframe to see what happens when a new signal appears. As expected, the fifth trade closes and a new set of five trades re-appears.
I will post the EA today for everyone to test. However, DO NOT GO LIVE with this EA yet as I have not incorporate the 200SMA filter.Hi
Just use 1 or 5 or some small number INSTEAD OF ZERO
OILFXPRO
Thanks for the clarification, OzFx. The reason why I asked is that I've downloaded the EA from Forex Factory and find that it doesn't do a short.
When digesting the code, I found out that the author has put something like if Stot<0, go short - thus the no-trade as the Stot value never goes below zero.
fxbs' explanation is much clearer if one's intention is to write the EA.
I am currently re-writing OzFx's EA from scratch with the money management feature as per his original methodology. I am forward-testing it now for bugs. Once it is certified as working, I will post it here.Hi
Are you going to post MQ4files or just ex4 files?If the latter will it have expiry date?
OILFXPRO
Hi Azmel,
Since the introduction of 200SMA we had 18/21 successful trade.
I think that speaks for itself.
Thanks again
OzHi OzFx,
That is a good ratio. What is the success rate without the 200SMA filter in place?
The EA that I have released above just uses the AC and Stochastic indicators for the time being.
Regards,
Azmel.
Broker selection
Oh yes, I forgot to mention - when selecting your MT4 broker, best find a broker that uses GMT (or GMT+1 at worst) as its internal server clock.
The reason is that we want the new daily candle to preferably appear during this time. Selecting a broker with a different clock will cause the candle to appear late, thus missing a good signal.
Thanks ozfx for this great indicator. It is simple to use and the most important thing .... profit.
My Backtested Results
I backtested from Julyo4 to Jan08 on a 10k account using 0.2 lots on one occasion and allowing the MM feature to manage at 20% equity on the other. Data is from Alpari - I'm unsure as to what time it would be based on.
I did get some error messages in the journal which I don't know what they mean. May explain the 80% instead of 90% modelling quality ( "OrderClose error 4108" & "unknown ticket 453 for OrderClose function" appeared many times)
I would say the results show promise but cetainly needs some refinement. Perhaps the 200ma filter will improve performance. The account is still intact after 3.5 years (460 trades) and the win loss ratio is 55.6%-44.4%. One obvious problem I see is the average loss is greater thn the average gain.
Well done for giving up your time and effort to get the project off the ground, and also for sharing. I'm sure I speak fo everyone "we appreciate it". Cheers, Jase.