Zeusjoes Market-Neutral Hedge Strategy

Also another thing, how many pips must one pair move in order to call it 'off' correlation?

Hello!
When i try to use OverLayChart -Point.mq4 i have the following error: zero divide. I have Ibfx with miniaccount demo.
Anyone can help me? I saw on the others that indicator work.

Maybe one thing... I think on IBFX your mini account symbol is GBPUSDm.... would that make a difference?? I use FXDD, and the symbols are just GBPUSD format, and the indicator works fine....

As to how much "un-correlation" to trade... it's dependent on the time frame and the pair's history. Look back at a given setup, at your chosen time frame and see how far apart they get, and go from there. Be sure to set a MaxLoss as well, or you could get hurt REALLY bad if they go out of correlation hundreds of pips, which can and DOES happen. But then, they will "re-correlate" and you can trade again at the new equilibrium. The key is to limit your loss, get out at a reasonable level, and wait for the market to re-establish itself..

With Chart below, I used a 15m time frame. Don't know if this was a good Delta or not, I don't trade the 15m... but going back on the chart with the indicator will give you an idea. So just for the example part, The left red line is the point in time at the candle close you have a Delta of 13 pips. (Sounds too low to me, but for the example...) at 01:45. You would have sold the GU at 9927 and Bought the EU at 5748. (Points 1 and 2) Then, the lines cross (or zero) at sometime just before 02:45 (right red line, Point 3). (If you were watching this in realtime, you would see when they touch and could close your positions just then, but since we are looking back, I will use the close values of the 0245 candle.)

So, close at GU= 9926, for a 1p profit... and the EU at 5760 for a 12p profit, total of 13 pips LESS your TWO spreads... so not many pips profit, but you get the idea.

Shorter time frames yield smaller gains, but you would have smaller MaxLoss risk as well, and could possibly trade more often.

the 1 hr time frame shoots for maybe a 90p delta, making about 90p in profit, but running a MaxLoss of 150 or so... (allowing for a total of 240p of un-correlation... )...

Sort of how I look at it...

If that is the OverlayChart-Point indicator, I've noticed that it repaints.
As the chart progresses in real time, the past overlay shifts up or down.
You can't look at past history on the chart and determine anything. When what you are looking at was current, it may have been aligned totally differently.

It's easy to see it repainting if you put it on a 1M TF.

I think that this best can be avoided by using a starting date Time in the past instead of number of Grids or Bars.

Thanks mrkam. I changed indicator settings from USDCHF in USDCHFm and work fine. Thank you, and sorry for my english.

Good point!! and Yes, is is a "repainter"... but from what I have seen, not significant for how I use it... Pretty much, I just want to know the past 30 days, how far apart, on the hourly, plus or minus 10pips.. Do they get 50, 100, 150p apart? How many times? stuff like that. Pretty much just to get a "feel" for price action movement....

I'll have to check that starting date idea?? Is that a setting in the indicator, or a revision you made to the code??

The things you want to see is exactly what it won't show.
Put it on a 1M TF and you'll see what I mean.
For example, I have seen the overlay be well above the price. The next bar it will completely shift to below the price.
I don't know what, if anything, you can see by looking back on a chart.

As for me, i take each day open price as a start point and look how many pips each pair makes from this point. On 15 min TF i place 2 orders when difference is greater then 20 pips. Then i set EA which close positions when profit is 15 pips.

1. I think there is a lot better way to trade these pairs than using a hedge. I think the hedge wastes profits.

2. If you're going to use a hedge,ZeusJoes' suggestion seems to work better than any indicator I've seen.

He said to use a demo platform and open up one position each on the two pairs, at the same time. You can watch the "profit" column to see disparity. I've played with three or four trades that way and have made profit on all of them. It may just be luck. I don't know.

Open up two longs (or shorts, it doesn't matter), one on each pair.

When the profit or loss on one of the pairs greatly exceeds the other, then hedge in the opposite direction.

Example- Open a long GBP/JPY and a long EUR/JPY.

Let's say that an hour later the G/J is - 600 and the E/J is -100.

Either the G/J is TOO low or the E/J is TOO high or both.

On your trading platform open a Long G/J and a short E/J.

Close them when you have the desired profit. Most of the the times I have done this and made a profit, the indicator was giving the opposite signal. I don't trust the indicator.

This isn't a complete system. It has to be determined how much of a disparity is enough to trade.
How much profit do we close them on?
When do we close them if they go against us?

Most of people waste all account trading one pair
Or you have super system?

If you think this hedge is any safer than trading a single pair, you are deluding yourself.

As you sad it's not a system. There is no particular rules. I try to use it different ways by now, and maybe i'll get some iteresting results. Who knows...

What is you favorite system at this moment? I'll be glad to know

I use a daily system- have been for a few years. It's a commercial system and I can't post it. It's been in a drawdown for a few weeks and just took off again a few days ago.

It's your commercial system or not? You can PM link to me if you whant. May be i'll be interested in it too. Thanks a lot!

For some reason the 1hr is not near as volatile as the 1m.... agreed...stay away from using the 1m!! But the one hour is comparatively pretty stable.

We all have the same questions.. and all we are trying to do is get on the right side of probabilities....

Entry: When? at what price divergence? (the original method uses the raw data, chart vs. chart. Others use indicators base on calculations to the second currency, all in an attempt to show price action against the base chart....)
MaxLoss: How Much? (enough to let the trade develop, not too much to take excessive losses...). THIS PART IS NECESSARY, or will blow accounts at some point. Ask FreedomRocks people!!
Exit: When? specific profit? at some sort of "zero point"?

All good questions, and all part of developing a system.

Using the one hour has been solid for me for almost a month now. Anywhere from 50-90p a trade, and the indicator works just fine... BUT! I use very specific MaxLoss (have an EA that tracks the net profit of two positions, and will close at a specific loss, or a specific profit. I use both to cover both sides of the trade.)

This might be more appropriately called a "semi-hedge", and as correctly stated, it is NOT inherently "safer" than a single position. Again, all probabilities, but "usually" it moves more slowly and not as far either against us, or for us. Just a different trade to manage, in a different way.

And great discussion!!

I wouldn't trade on 1M, but with a repainting indicator, the 1M speeds up the action so you don't have to watch it as long to prove it repaints. A repainter, repaints as bars are added, on the 1H TF you have to wait an hour for a new bar.

I don't believe there is any way this indicator is more stable on 1H than it is on 1M. An indicator is either programed to allow repainting, or it is not. It just takes longer screen time to see it on 1H and it appears to be more stable.

I'm not saying the indicator won't work if you are basing trades on it's present position (although I have my doubts), all I'm saying is that looking back on a chart, it's impossible to tell how the indicator really looked in the past. It repaints the past.

Hi mrkam!

What distance do you use on 1H to enter and what is you target profit and target loss?

OK, last post on this one... but here you go

1. Regarding repainting. I have seen the 1m move, and I have seen the 1H move, but only once, and only within the first few ticks. I don't know why. But once you set it up, and let it run, the past info IS valid.

Seems like you have to keep the zoom level the same, or else things will change... I use one zoom less than max.... I know,... seems like a pain, too much to keep the same, but if you do, it works...

But don't believe me. Do this. (I have). Each hour, record the close value of the main chart (line). Record the Value3 of the indicator. Start with the closest past closed bar, and record that closed bar along with every proceeding closed bar. You will see that the values remain the same. That is, it does not repaint, that is, it is accurate.

that's all on that.

2. Here is a trade for today. I took it because of the big delta after a pretty flat time. Easy 48p, in 3 hrs, after spreads.... Great example of what this is... The GU takes off up, while the EU stays down, then the EU takes off up to catch up, eventually hitting the GU... that is what we are doing here, playing the time lag give and take between two pretty dang correlated (over time) currencies...

Delta at close of 1000 was points 1 & 2 (delta 62). GU was above the EU, so Sell GU, Buy EU. Closed at cross, point 3. Since we are dealing with closes, you can look at the charts and get your values. GU was -2, EU was +56, net +54 less 6p for both spreads, +48 in the bank.

I am working on an EA that will open automatically at a set Delta, not wait for close. It will also close at whatever Delta you set, for me, zero. But it is still in development. So, for now, I have an EA that will close all positions at a set profit or loss level in pips. So i set it up on two charts, one at my set profit, one at my set loss. For the GU/EU, I usually look for a delta of 90p. I took this one at less because of the recent price action. I usually set the Maxloss EA at 2X the Delta. Here, at delta 62, I set at -124p. Since the two pairs are already at 62 p uncorrelated, they would have to get to 186p uncorrelated to hit my MaxLoss. Fine. I am willing to risk that because 1) it is rare (based on back checking price action, hasn't hit 2x in over a year...) and 2), (as BillSS has stated, the two pairs WILL go hundreds of pips out.. at some point, but very very rare.. so, we NEED to get out if this happens...).

OK.. rambling, but like I said, last post here....

So, I set the MaxLoss EA at 2X the delta, and the profit EA at the delta less spread... in this case 62-6, or 56... which hit, plus or minus a couple of pips for closing...

I know BillSS says the indicator I use is worthless, but I have documented otherwise. (and at over 1700 posts, I am not going to argue! But to be true, I am right up there, different forum, but that is neither here nor there...!!) It is tedious, but do the work youself. You HAVE to verify any system yourself. Don't believe me, or anybody. Record each close and each value3 for every hour starting now... you will see if it changes. Give it an hour or two to settle first. Don't know why, but this seems to make a difference. I use a virtual server, so my trading platform is on 24/7, undisturbed.

You CAN use the past info. Check the indicator with actual charts, using closes. Check as far back as you want, same process as above. Check at the larger deltas, check at the crosses, do the math... it works. Based on doing this,... the GU.EU hit 8 trade in June at 90 delta. Those 8 all closed at delta zero for a total of 750p (less 48p spread...). Max unrealized DD for any open trade was -90p. Not once in June hit a loss.

This is not rocket science. Simple price action of two currency pairs, with an indicator to show entry & exit.

Good success to all. Hope this helps, but this discussion seems to be going in circles, meaning what can be said has been said (so far!!).. Bye.

correct he if i'm wrong here guys, but i think kurkafund is doing something very similar on his triangular arbitrage page