Hedge and Correlation Strategy | Page 151 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Correlation + Momentum Strategy? 6 replies

MT4 MTF Correlation table and Correlation oscillator indicators 155 replies

InsomiaFX Correlation Double Hedge EA 66 replies

Looking for EA for Correlation Strategy 6 replies

Hedge strategy (not a hedge fund) 13 replies

Attachments: Hedge and Correlation Strategy
Exit Attachments

Hedge and Correlation Strategy

  • #3,001
  • Mar 18, 2011 10:17am Mar 18, 2011 10:17am
Quoting Indrek
When trading at serious level (full-time) then it is obvious that you diversify risks.
I never hedge too many positions and I always trade different systems at the same time. And I never open any trades for getting just 10 pips.
Current drawdown of one part of your open trades should not be important or worrying, otherwise you are overtrading, or your money management is just too bad, or your account is far too small.
If you never open a trade for getting just 10 pips, then you have to use higher TF, not M5 or M15.

Now I understand why DD of 200 pips is nothing for you.

This method at a smaller TF is for scalping, but for higher TF is for long term traders.

I think that using this method for Higher TF is much safer.

Good Luck lol.
  • #3,002
  • Mar 18, 2011 12:36pm Mar 18, 2011 12:36pm
Quoting Dreamliner
Hello Jack-fx,

I have by no means disappeared, am still here and still trading this system with success. However, the thread has taken a turn which I do not like. People have begun posting all about the EA. I absolutely detest EA's and believe they are the blight of Forex Trading. So I simply don't join in on that discussion.

Have a great day.
Thank you for your kind reply.
I also perfectly agree with your position and join to the group of "manual" traders
You know why? I've been one of the EA researchers, for so long.
But now, for the first time, i find a way of trading which makes me feel very confident when I trade!

With all respect to everybody (EAs or not), if a logistic solution is needed, I think the most natural one is that the "EA school" would migrate to a new thread -after all Steve Hopewood already tried to.
Again, I'm not addressing to people, but to the mere subject (automation of this strategy).
In fact, focusing on EAs has caused the fact that many brilliant collegues moved away.

That being said, I think this strategy is really robust and deserve a lot of attention.
So, just as an example, there are a couple of points worth to delve into:
- risk management (spinfoil, where are you??? )
- possible statistical improvement


Great trading to all
  • #3,003
  • Mar 18, 2011 2:16pm Mar 18, 2011 2:16pm
Quoting speedytina
Dreamliner...if you are reading this post, I have a question. I note that you often warn against trading negative correlations. Could you explain in more detail why this is so? I don't understand why the risks (drawdowns) should be any greater than trading positive correlations. Since all the discussions have been about positive correlations only, I'm not even sure if the same indicator rules (entries and exits) would apply to neg correlations. I read somewhere that negative correlated pairs should both be shorted. Is this true? How would one use...
I also had problems with negative correlation but finally I studied more statistics and theory of correlation and found where the problem was.

First - in stochastic diff settings you have to invert negative correlated pair. Otherwise you will had inverted signals.
Second - for example in EURUSD - USDCHF what is really correlated is EUR and CHF in their behaviour against dollar. Correlation is negative because in USDCHF dollar is in the first place.

So - if you buy EURUSD - you buy euro against dollar so the other part of your position should be sell CHF against dollar - so also buy USDCHF.
With negative correlation You do buy buy or sell sell trades with direction depending on positive or negative standard deviation of correlated currencies.

To make it simple - if you have EURUSD above USDCHF in stoch diff - you sell EUR and buy CHF (so also sell USDCHF).
If you have EURUSD below USDCHF in stoch diff - you buy EUR and sell CHF (so also buy USDCHF).

It sounds weird - but is correct with correlation.
If you sell EURUSD (you buy USD) and buy USDCHF(you also buy USD) you risk that with weakening dollar gap will be growing till it blow up your account.

And I think main problem of EA is a problem with standard deviation. EA is giving signals on calculations made in pips , I guess, while standard deviation is a relative value.

I can be wrong - I do not trade with EA - but I trade on correlation based more on statistics - and with volatile moves it happen a quite few times that EA was giving signals when deviation was close to zero or it was still rising so the gap was about to grow further.

StDev below 0.2 means that in relative values there are really no gap (but of course there can be big gap in pips because a volatile moves in both pairs and their different average volatility).

And I think that is why EA made problems last week.
Trading manually on calculations of stdev was ok. - more volatile than usual with bigger gaps but without being trapped in growing gap.
  • #3,004
  • Mar 19, 2011 4:25am Mar 19, 2011 4:25am
Quoting hblekot

StDev below 0.2 means that in relative values there are really no gap (but of course there can be big gap in pips because a volatile moves in both pairs and their different average volatility).

....
Trading manually on calculations of stdev was ok. - more volatile than usual with bigger gaps but without being trapped in growing gap.
Hi hblekot,

what do you mean when you talk about stdev?
It's exactly what I'm looking for.
It isn't part of "Stoch. different pairs 1.7" indi.

Maybe are you talking about "Basket profit 2 pairs" indi?
I show it in the central part of the attached graph.
I'm still trying to correctly use it. (Especially) on hourly graph it can provide us with signals even when the 'traditional' "Stoch. different pairs 1.7" indi doesn't.

That's what I meant when I wrote about possible statistical improvement of this strategy.

A feedback from everybody else who likes or use it is welcome!
Attached Image (click to enlarge)
Click to Enlarge Name: EURGBP-EURUSD 1H.PNG Size: 96 KB
  • #3,005
  • Mar 19, 2011 7:10am Mar 19, 2011 7:10am
Quoting jack-fx
Hi hblekot,

what do you mean when you talk about stdev?
It's exactly what I'm looking for.
It isn't part of "Stoch. different pairs 1.7" indi.
Stdev - is just a standard deviation.
I do not have any indi for it.

Here more theory:
http://en.wikipedia.org/wiki/Standard_deviation

Do not bother with calculations - what is important is just mechanism of deviation. As You can see there on graphs with normal distribution most of the time stdev is below 0,2 and rarely goes above it and very rarely goes around 0,5

You can use clusters indicators to know the value of deviation
http://articles.mql4.com/484

CCFp - shows absolute values best to use for standard deviation. Choose in setting EUR and GBP - and on graph you can check deviation between them. When value for EUR is 0.0010 and for GBP -0.0010 the deviation between them is 0.2%.

You can evaluate when is a good moment to enter the position.

But remember that problem of growing gap is still present - you will know that there is a deviation but deviation can rise widening the gap more so averaging position and proper money management is a must.
StDev can rise to 0,5% and on some extreme situations even to 0,9%

But when stoch diff is giving you a signal and deviation is very low - for sure it is not a good trade setup. The gap is about to rise and it can rise the way you open your positions and you will make a fortune or it will go opposite way and you will be losing on both sides of the trade.

It sounds complicated - but the way it works is quite simple - standard deviation is just a divergence in correlation and that divergence will be your profit when both sides of the trade goes back in balance.
  • #3,006
  • Mar 19, 2011 9:13am Mar 19, 2011 9:13am
Quoting jack-fx
Thank you for your kind reply.
I also perfectly agree with your position and join to the group of "manual" traders
You know why? I've been one of the EA researchers, for so long.
But now, for the first time, i find a way of trading which makes me feel very confident when I trade!

With all respect to everybody (EAs or not), if a logistic solution is needed, I think the most natural one is that the "EA school" would migrate to a new thread -after all Steve Hopewood already tried to.
Again, I'm not addressing to people, but to the mere...
Hi:

I also wonder about 'dreamliners', this thread starter, not posting. Now I know his reasoning, I completely understand his position. I guessed as much...

For me, whether it is EA trading or manual trading, I like the concept of this strategy. However, I feel also same as dreamliner said, this thread was taking to the different direction than what he is intended when he start the thread. Now, people discuss and show great interest about the EA then discuss more about the concept of hedging/correlation strategy.

Example: if trading EURUSD and USDCHF, trading on opposite direction on these 2 pairs can make money quickly or your account will be in huge draw down. What I wanted to see the discussion on this thread was , in this case, how we could reduced the draw down. By adding more lots? if so, what direction and which pair? or could be better to introduced EURCHF (3rd pair to the basket)? if so, which direction and how many lots to put on? when to add the 3rd pair?

Now, I see where this thread started as manual trading, now turned into EA trading. I see some members wanted to discuss these issues time to time, but next members come to the thread and post something about the EA not functioning. Before you know it, thread is moving to discussion of EA's malfunctioning. Don't give me wrong. I like to use EA to trade and I believe using EA to trade can help traders in many ways if use it correctly. HOWEVER, when strategy is prematurely develop and using the EA is not a good idea. So many EA trading fails because EA was written without the solid trading plan.

Nonetheless, I am checking the thread posting day to day hoping more members really talking about improving strategy. I absolutely love custom indicator(s) to alert me, so that I can manually make my trading decision. I wish to see it more of it on this thread; more discussion about how to improve such a good trading concept into a super robust trading system. I am also new to this forum, so I am sure this post will not be posted until Admin's approval. Hope to see dreamliner could come back and star posting more about the concept as he intended from the day one.
  • #3,007
  • Mar 19, 2011 3:45pm Mar 19, 2011 3:45pm
Guys, for your consideration ( I am trading this way 1 week with absolute no loss ) ( I had to buy a custom indicator but I am sure here are good coders that can reproduce it ):
If we begin with correlations in 1140 why not watch un-correlations in 5 min ?
1.- Watch correlation > 80 in Daily in 200 days
2.- Watch correlation < 80 en TF 5M in 200 periods:
* If correlation < 80 enter 1 lot, 1 sell, 1 buy according FXCM TradeStation visual mode.
* If correl < 85 ( grap grow up ) enter a second lot
* if correl < 90 ( grap continue up ) enter a thrid lot
3.- Take profit a 1/2 of "separation" of gap in visual mode fxcm tradestation
4.- Optional check if spread is < 10 .

In graphics I show a operation with 50 pips profit

We need a indi that show us 2 correlations, Daily and 5 min . Somebody can code it?
Attached Image(s) (click to enlarge)
Click to Enlarge Name: Correl_ff_3.PNG Size: 97 KB
Click to Enlarge Name: Correl_ff_2.png Size: 34 KB
Attached Image
  • #3,008
  • Mar 19, 2011 5:47pm Mar 19, 2011 5:47pm
Quoting AlbertoFX
Guys, for your consideration ( I am trading this way 1 week with absolute no loss ) ( I had to buy a custom indicator but I am sure here are good coders that can reproduce it ):
If we begin with correlations in 1140 why not watch un-correlations in 5 min ?
1.- Watch correlation > 80 in Daily in 200 days
2.- Watch correlation < 80 en TF 5M in 200 periods:
* If correlation < 80 enter 1 lot, 1 sell, 1 buy according FXCM TradeStation visual mode.
* If correl < 85 ( grap grow up ) enter a second lot
* if correl < 90 ( grap continue up ) enter a thrid...
Hi,

A very good idea, but can you or somebody else can think of another way to know which pair is BUY and which pair is SELL, execpt the use of FXCM TradeStation visual mode?

This will be very helpful.
  • #3,009
  • Mar 19, 2011 7:04pm Mar 19, 2011 7:04pm
I not know how code it, but meter the separation the EMAS of pair1 - EMAS pair 2 and find valid "gap" value then we can enter.
Attached Image (click to enlarge)
Click to Enlarge Name: Correl_ff_4.png Size: 97 KB
  • #3,010
  • Mar 19, 2011 7:19pm Mar 19, 2011 7:19pm
I readthe trade theories here and like them. But why not just hedge same pair with 2cd account and set pending hedge order outside parameter of your trade?
  • #3,011
  • Mar 20, 2011 12:42am Mar 20, 2011 12:42am
Even, had you used this method?
First problem is to have 2 accounts.
How is "set pending hedge order outside...." ?

Thanks...


Quoting Pips-and-Dip
I readthe trade theories here and like them. But why not just hedge same pair with 2cd account and set pending hedge order outside parameter of your trade?
  • #3,012
  • Mar 20, 2011 11:01am Mar 20, 2011 11:01am
We need a EA for trailing stop/profit this way:

1.- If profit reach 25 pips then "lock" 20 pips
2.- If profit go againts such 20 pips then close both pairs
3.- If profit go 30 pips then "lock" 25 .
4.- If profit go 35 pips then "lock" 30, etc, etc...

With this EA is more easy follow hedge pairs profit and not put fixed profit

Do you know some easy EA for manage this?
Thanks..
  • #3,013
  • Mar 20, 2011 4:06pm Mar 20, 2011 4:06pm
Quoting AlbertoFX
We need a EA for trailing stop/profit this way:

1.- If profit reach 25 pips then "lock" 20 pips
2.- If profit go againts such 20 pips then close both pairs
3.- If profit go 30 pips then "lock" 25 .
4.- If profit go 35 pips then "lock" 30, etc, etc...

With this EA is more easy follow hedge pairs profit and not put fixed profit

Do you know some easy EA for manage this?
Thanks..
Hi Alberto,

Take a look at this thread: http://www.forexfactory.com/showthread.php?t=89371
it seems to me that this EA could easily do what you are looking for.

By the way Ive been playing with this strategy with mixed results, biggest problem seems to be DD, your observations are very interesting, looking forward for your ideas.

Saludos!
Roberto
  • #3,014
  • Mar 20, 2011 4:41pm Mar 20, 2011 4:41pm
Hi.
I will follow this 8 pairs this week. Also I will publish possibles call alerts with graphics.
Attached Image (click to enlarge)
Click to Enlarge Name: Correl_FF_Pairs_D1_20032011.PNG Size: 30 KB
  • #3,015
  • Mar 20, 2011 4:49pm Mar 20, 2011 4:49pm
Thanks
  • #3,016
  • Mar 20, 2011 4:50pm Mar 20, 2011 4:50pm
Here is possible trades in near hours.. may be in 2-6 hours..

Thanks
Attached Image (click to enlarge)
Click to Enlarge Name: Possibles_20_03_11.PNG Size: 46 KB
  • #3,017
  • Mar 20, 2011 10:17pm Mar 20, 2011 10:17pm
Nice gap forming in GBPAUD and GBPCHF . Waiting until < -80%
Attached Image (click to enlarge)
Click to Enlarge Name: Correl_ff_5.PNG Size: 103 KB
  • #3,018
  • Mar 20, 2011 11:44pm Mar 20, 2011 11:44pm
Trigger was -80, entering 1 lot
Attached Image (click to enlarge)
Click to Enlarge Name: correl_ff_6.PNG Size: 102 KB
  • #3,019
  • Mar 20, 2011 11:48pm Mar 20, 2011 11:48pm
Hi, AlbertoFx,

I am new to the thread. thanks for sharing, where did you get the custom indicator? thanks.



Quoting AlbertoFX
Guys, for your consideration ( I am trading this way 1 week with absolute no loss ) ( I had to buy a custom indicator but I am sure here are good coders that can reproduce it ):
If we begin with correlations in 1140 why not watch un-correlations in 5 min ?
1.- Watch correlation > 80 in Daily in 200 days
2.- Watch correlation < 80 en TF 5M in 200 periods:
* If correlation < 80 enter 1 lot, 1 sell, 1 buy according FXCM TradeStation visual mode.
* If correl < 85 ( grap grow up ) enter a second lot
* if correl < 90 ( grap continue up ) enter a thrid...
  • #3,020
  • Mar 21, 2011 12:53am Mar 21, 2011 12:53am
gap grow up < -85% then I add another lot:
Attached Image (click to enlarge)
Click to Enlarge Name: correl_ff_7.PNG Size: 111 KB
4 traders viewing now