Hedge and Correlation Strategy

Tokyomaster, I'm not familiar with IForce. Care to share?

These charts are nothing more than one currency superimposed on another. The scales are different & they only APPEAR to be one on top of the other. We asked the MT-4 program to do that for us & it does exactly that.

The ILLUSION shows tendencies of each currency relative to their correlation to each other. They may not really be one one top of the other. Nor may their respective ATR be even close.

Wheather these pairs are seperated, crossing, or converged is only an ILLUSION. But paints a clear picture of their relationship to each other.

We're trading the tendency here, and the likelyhood that that tendancy will continue & repeat.

We can normalize the pairs through lot size, so that movement, of anticipated movement is equal. For instance the ATR difference between EUR-USD & USD is generally known to be 1.5 : 1. But yesterday I figured the difference at 1.26 :1 & traded accordingly & successfully.

Take EUR-JPY & GBP-JPY, with a 85% correlation factor. EUR is 1.4 to GBP. Look at EUR-JPY & AUD-USD. CF is 89% But the ATR factor is something like 4:1 or something like that. We need to use our position sizing such that we can establish a level playing field & double our chance that these pairs are going to do what we need them to do within the time we want it done.

I think the key to these charts is to always remember that they aren't the truth, but are representative it WHEN PRESENTED IN RELATION TO EACH OTHER. We can't just blindly trade from the superimposed charts. We need to do a lot of other work to justify our intended action. The superimposed chart is merely the PRIMER. Get's our attention. We need to do the rest.

For me, this is no different than trading Index Option spreads from 20 years ago, so, once I cast my eyes on this chart superimposing indicator, I was good to go in about 3 minutes. And there's no time decay, no delta, no nothing. A pip is a pip is a pip is a pip. Can't get any better.

Now if only Barack the Ever-so-Benevolent doesn't give away the ranch more so than he already has at the G-20. He could give them all of the US cash....it's worthless anyway. Thanks for nothin Barry...you too GeorgeII.

It would be just about right that I finally wrap my arms around a reliable trading method & learn how to use it & profit from it when no sooner comes the US Forex Cops to raid my house & declare martial law for the sake of a Green Planet & of course for the Children & force all Forex Traders to go get a real job so that all the layabouts can afford more beer cigaretts & lottery tickets. Costa Rica looks better & better all the time!

If you have high correlation on the OPEN prices, on a daily chart for example, and your HIGH and LOW price correlation is not strong(or is opposite of the desired correlation), this is where you are going to see the draw-down. Since each candle has a structure, i.e open/high/low/close, measuring only 25% of that structure, in my opinion is incorrect. This becomes very important with the highest time-frames, with smaller time-frames this is not as important, as each candle is rapidly changing. But on a 4hr, daily, and weekly, since each candle is updated on a slow scale, measuring only the OPEN correlation, opens your account to the risk of draw-down with HIGH and LOW correlation.

where can one get the correlation on O,C,H L?
Thanks

This is a question that I had. So, the crossover means nothing, right?

For me what this method needs is an indicator that will show the divergence between the 2 pairs according to pip count.

In other words, every time the pairs cross it begins the count. The target is what ever I decide it should be, and that will be according to the TF. The smaller the TF, the smaller the pip count number will be.

Let's say I'm using the H1 TF, and that my research has shown that 75 pips is an adequate divergence for that TF. So I set the indi input for 75 pips.

So, once the pairs cross the count begins. If they ever diverge by 75 pips I will be alerted so that I may set up my trades. My entries and TP will be done manually, but with 75 pips (or what ever) between the 2 pairs, I am sure to get some "snap-back" on the price action.

Does this make any sense to you indicator builders? I can see it very clearly in my mind, and I think it would be fantastic. Let me know if I need to clarify anything please.

I think I've got this figured out.

Friday I put on a 3-way hedge (GBP-JPY, EUR-JPY, EUR-GBP) & as I was easing out of my hedge, successfully, the damn platform closed on me. I was still exposed 4 positions short GBP-JPY. Couldn't get out & GBP-JPY headed straight north & then closed later.

I'm still exposed GBP-JPY, 4 positions, 40+/- pips to the bad.
I'm going to choose a different pair with which to hedge this out.

Unortunately, I've got to go take my mom-in-law to Palm Sunday Service.
When I get back, I'll write everything I do to set this up & post by around noon EST. I think you folks will understand more clearly then the thought process & methodology I'm using.
I really think this has legs & I'm psyched!!

Lets see if this helps out any.

Probably just confuse the issue......here goes:


OK.

Earlier I had spoke of the need to mitigate 4 GBP-JPY which I was locked out of on Friday due to the platform shutting down, leaving ne now with a net 45 pips (X 4) to the negative.

EUR-JPY and GBP-JPY are about as converged as they’re going to be, as I had suspected as wind-down for the weekend approached.

I could just watch GBP-JPY, wait for a pull-back, buy into it & hedge out. But this is GBP-JPY. It could do anything. So I’m going to try another convergence hedge. See what happens. It’s only a micro account!

I’ve chosen GBP-JPY and AUD-USD. The correlation factor as of Friday is 72%. Not the greatest but good enough.

Take a GBP-JPY chart & Overlay onto it a AUD-USD. GBP-JPY is the pair with the highest ATR, therefore this one should be the one doing all the moving around. Weather this theory holds water remains to be seen.

First call-up a daily, 1 hr. and a 5 minute. Do these overlays resemble each other at all. Yes. They do.

We’ll set this up on the Hourly, but monitor the trade on the 5 minute.

Look at ATR(20) and ATR(5), on the hourly. GBP=JPY is 384/299 and AUD-USD is 175/134.
Divide the larger by the smaller. Gives the ratio comparison of relative robustness of each pair compared to the other. A volatility comparison of sorts.

Now, you need to scale these pairs. GBP-JPY scale is on the right. AUD-USD is on the left.
Go to the 1 hour & zoom out as far as you can. Get the read of the highest high, either pair, and the lowest low that you see on the screen. I got GBP: 148.77 and 127.13 AUD: 0.7164 and 0.6335.

Now go to the 5 minute and do the same. I got: GBP: 148.81 and 140.53 AUD: 0.7230 and 0.6870

Differences in range is on the 1 hour: GBP : 2164 AUD: 829.
Divide again. Ratio: 2.61

Do the same on the 5 min.. I got ratio of 2.30

These numbers are certainly in the ballpark. Not scattered all over creation. So, I ascertained this set-up to be a “keeper” & I’m going to with it.

The ratios from the scale comparison is now your position size index. I took 2.61 and 2.30 and split the difference because I’m a fair guy. So, GBP-JPY position is 1 unit to 2.5 of AUD-USD.

Now, from here on, don’t pay attention to the AUD-USD scale over to the left of the screen, only to the GBP-JPY over to the right. The GBP-JPY is the actual screen for real. The AUD-USD is only superimposed over, therefore it’s scale is no longer relevant.

These pairs, we're looking for them to converge, overlap, and cross. This doesn't really happen. We're watching, representative price action, of 2 different pairs, on 2 different charts, as they inter-react with each other due to their correlation. We can see how one moves higher than the other, etc. it's Tendancy RELATIVE to the other. Don't get lost watching that scale. You have done your due dilligence before you pulled the trigger on the trade. Now just watch the bars of the 2 pairs..

Also, we are only interested in how much the GBP-JPY and the AUD-USD approach each other on the chart. The actual numbers for price on the grid isn’t important. Only the distance (in pips) between the 2 pairs.

So, I’m going to wait till 0:00 GMT Monday, see what’s happened for Sunday, to eliminate surprises, and just enter at market 10 AUD-USD to offset my 4 GBP-JPY shorts.

We’ll see. Good luck everyone. [IMG]file:///C:/DOCUME%7E1/Peter/LOCALS%7E1/Temp/msohtml1/01/clip_image002.jpg[/IMG]

download the indicator for MT4, here:
http://codebase.mql4.com/4971

You would have to setup (4) indicators, one for each time series.
OPEN
HIGH
LOW
CLOSE

Is this thread still alive?

Chamane

Because I am making many pips with it so far and I want to make sure there is no major reason not to use it. Also the issue of no stop losses scares me somehow. I am using 5min charts.

Chamane

Hi Chamane,

This method puts on the comparative pair chart & sizes it to suit the screen size of the original chart. The ATR is scaled in that manner as well as the time frame size.

So, if you are successful with this, it's coincidence only. My first few trades live with this were home-runs. But I discovered that ATR must be scaled as well as position size...which is fine. Then I kept getting on the wrong side of trades & discovered that I was just seeing an illusion. In fact, in some instances, different time frames saw the pairs switch positions. That was my first clue which I refused to believe. Give away a few hundred pips & it got my attention.

But, the charts adjust as time progresses to accomodate relative overlay. So your graphic representation as seen & analysed at one time won't be the same at a later time because of the accomodating adjustment. There is no anchor common reference between the 2 charts except a visual illusion. And that's precisely what you are seeing...a illusion which constantly changes.

Try this site:

http://www.forexfactory.com/showthread.php?t=160679

Billbss is onto something. He threw this up for comment, etc. & there's been a lot of work put into refining this and maybe getting it to work reasonably well.

I think this approach is solid as options are often traded in a L/S betting on the spread (straddle / strangle). I suppose that one could get a fix on correlation factor in a excel program, in fact I'm sure you could. But seeing visually on a chart is so much quicker & easier. A picture worth 1000 words, etc.

Good luck.

yap i remember this..and the other thread by Zeusjoes....
i seriously think we found the good trading system until sudenly everything start to go wrong....
mostly because the volatility cause too big gap between those pairs that couldnt be handled by ordinary margin...

Dreamliner..pls take a look those thread and see if there is something that you can add...i still hoping there is somebody that could perfected that system...

petero,

Thanks for answering. I knew it was too good to be true, but maybe there is something to do about it. Right now, I am trying different things like verify if the pairs keep the same position in the 30 min and 5 min timeframes and trade the 5 min.

I also enter only if CCI is oversold for one pair and overbought for the other at the same time (wich is not always the case) followed by Heiken Ashi red and blue respectively at the same time. Maybe that should keep me out of trouble as long as I have stop losses. The relative position of the two prices is visually repainting as you say, but the strong negative correlation remains and that's what I should be building on. I will surely have a look at the link you gave me.

Thanks.

Chamane

correlated hedges work well when they work well. I had an eurtry/usddkk hedge that worked great for about 6 months, trippled my account and then markets flipped, after looking through history a little more indepth i discovered that for that particualr correlation the correlation flips form negative to positive and back every 2 years or so. something to consider if you really want to keep pursuing a correlated hedge like this, as soon as markets show signs of change, bail! keep an exit plan handy.

will it help if one takes a trade on a lower timeframe in the direction of the 4hr and daily ?

is there some kind of contingency plan from Dreamliner like if things go wrong, is one going to double up or whatever ?

it's a major disaster if you can win on 10 trades and then the 11th trade destroys your a/c.............

Billbss's pair divergence method at least has a stoploss.

Hi folks, has there been any more progress on this thread? It really feels like we are close but the interest seems to have died off somewhat.

Cheers.

I'm doing wonderful with it!

That's good news. Can you please give a rundown of the system you're using and indi's, timeframe, SL, TP, strategy etc...


Cheers.

Overlay trading is a mis-conception.

There is no gaurantee that the two currencies will ever recross back to be able to make a profit as prices get re-adjusted. Also, the gap changes on a daily basis.

Back test it and you will see it fails. And, yes you can back test trading 2 currency pairs, there is a thread on here which shows you how to get ovver the current limitation in MT4.

For it to work, you need to factor in trend/power/news.

Sure can, page 1, post 1.