FINEXX Correlation System EA - page 11

To add comments, please log in or register
Michael B
3377
Michael B  

I can confirm that all versions of this do not work in IBFX.... expired or zero divide errors...no mq4 file available...do not waste your time

ES

P.S. I am using build 217

tgee
3
tgee  

Expired

tbreaker:
Version 2.2 higher Stoploss gives better results but risk is also higher

Hi tbraker,

I'm new here, I already attached FINEXX ver 2.2, but it's not run, with alert that the EA already expired.

How can I have the new ver 3?

Or it's really can't run on FXDD and other server?

Thank you

dirtybrown
3
dirtybrown  

Correlation EA

I am trying to write an EA to do the following:

For 1-HR timeframe:

1. 15 period MA cross above 30 period MA on eur/usd

2. 15 period MA cross below 30 period MA on usd/chf

3. 15 period MA cross above 30 period MA on eur/chf

condition

if buy signal on eur/usd + buy signal on eur/chf + sell signal on usd/chf then execute buy on eur/usd.

TakeProfit is 1X ATR value

StopLoss is 3X ATR value

close half position on 1x atr, trail stop remaining balance

The person who authored this method claims an 85% success rate. -

I am new to MQL so if anyone can help or provide a sample EA close to this that would be great

__________________

EricJohnson
3
EricJohnson  

choosing pairs

Hi,

I have been trading major pairs that are over- layered as SMA's charted in a single lower window . I made my own indicator, but I think I saw somebody posted one similar on another thread.

The assumed inverse correlation of the EUR/USD and the USD/CHF or any two pairs seems a weak assumption when they are seen historically even on 30 minute bars.

I only take pairs that are on the opposite extremes of the lower window. I use Metatrader crosshairs to make a click and drag measurement. I have historical preset for the measured difference between two extreme pair SMA lines.

This is mentioned because I am working to make a (MACD style) trigger that would auto measure the separation of the lines . Then make an EA for trading the extreme pairs toward each other. I am a beginning programmer and do not know if this is a possible, due to varied values being crammed into a single window. Of course any input would be welcomed.

Another note, I do not ride these pairs to crossing of the SMA's. I take profit at a good pivot or support level of one of the pairs. I just jump to the next most extreme pair. I can see when a pair is stretched on a regular hourly bar chart, and it is confirmed on in the SMA window. The SMA is a nice way to get a market overview on a single chart, and see the relative angle of the trend.

Many times I just take single sided trades when a stray pair wonders far from the pack, like the USD/CAD does often.

I have read some of the criticisms of using the SMA overlayed window. There are scaling issues. If you keep in simple and do all of your historical measurements in like it was a real time progression, it gets quantifyable. That means look at the intersection of area measured, and the right margin of the chart.

Anyhow I will post graphics and indicators if somebody is interested. I have not been trading this for long, but it seems very successful. Also I know the difference between trading advice and a system. I apologize in advance for not likely specifying everything. I phase in and out positions, and look at well marked hourly bars with envelopes on them of guidance.

New here, nice forum, and good to be a part of the effort.

Yuri Predborski
561
Yuri Predborski  
dirtybrown:
I am trying to write an EA to do the following:

For 1-HR timeframe:

1. 15 period MA cross above 30 period MA on eur/usd

2. 15 period MA cross below 30 period MA on usd/chf

3. 15 period MA cross above 30 period MA on eur/chf

condition

if buy signal on eur/usd + buy signal on eur/chf + sell signal on usd/chf then execute buy on eur/usd.

TakeProfit is 1X ATR value

StopLoss is 3X ATR value

close half position on 1x atr, trail stop remaining balance

The person who authored this method claims an 85% success rate. -

I am new to MQL so if anyone can help or provide a sample EA close to this that would be great

__________________

Small bit of critics hasn't killed anyone yet, hopefully if won't kill a trader in you either.

Lets take 85% win as an example. Out of 100 trades 85 bring, lets say, 25 pips profit (20 pips takeprofit and 5 pips from trailing), 15 bring 60 pips loss.

This example gives us 2125 pips profit and 900 pips loss, leaving us with 1225 pips profit. Apply spread - 200 pips for 100 trades on E/U, 1025 pips left. Apply slippage and requotes, 200 more pips, very approximate. 825 pips left. Apply swaps (not all trades reach TP/SL within the same day, but considering TP is small, lets assume most trades aren't left overnight), -75 pips (again, very approximate).

We're left with 750 pips profit per 100 trades. Under perfect conditions. If win % drops to 80, we might lose at least half of our profit.

The best part about this strategy - you can't test it using backtest (test on historic data), MT4 backtest limitation - using several currency pairs at the same time is not possible.

If you still want to make an EA, I recommend doing it in a few simple steps:

Make an indicator that draws an arrow on the chart if conditions for this pair are met (MA 15 crosses MA 30, for example).

Then add a second arrow for another pair. Then add a third arrow for the third pair.

Then hide all arrows and make a check to see if there are 2 or 3 conditions met and put an arrow in that place. If it works, you can use the code for an EA. If it doesn't, it will be easier to figure out why it does not work in an indicator than in an EA.

Hope this helps!

PS: you should end up with something like the attached picture. All 3 arrows on the same bar. Spoiler: it ended up at stoploss.

I'm not providing any samples because:

a) I don't want to work for free and

b) you already have the best sample - MACD sample. If you really want to be a programmer, open it up, edit, save under a different name and start learning! If you don't want to be a programmer, be ready to pay a small fee for your programming needs.

Files:
To add comments, please log in or register