Hedge and Correlation Strategy | Page 106 | 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

  • #2,101
  • Feb 2, 2011 8:13am Feb 2, 2011 8:13am
  •  Ckp
  • | Joined Aug 2010 | Status: Master ur EXIT. Entry is secondary. | 29 Posts
Hi DreamLiner

Ive been trading correlations for 6 months with good success and would like to share some important points that have really helped me to avoid losing trades; Time Frame and Pair Choice.

Time Frame:
In my experience M1 and M5 are just too short and consistent profits will elude you.

Ive been trading M15 and M30. If I get a signal in M15, i try to compare the same signal in M30. If M30 is still a long way off from a signal, I wait longer until I feel comfortable or until both M15 and M30 give a signal. Over time you will get a feel for the pairs and their movements and when to enter. M15 and M30 give regular signals with acceptable draw down.

H1, H4 and D1 take a lot longer to give signals and draw down can be significantly more. In a nutshell: Less signals, more stable, more draw down, higher profit per trade, BUT less trades.

Pair Choice:

Don't trade negatively correlated pairs! Not worth it! Believe me I've tried.

Trade only positively correlated pairs with a total of 3 different currencies one of which "MUST" be hedged ie: (Buy)EurCad (Sell)EurUsd. This way you are hedged for any risk event on "EURO". Risk events for "CAD" or
"USD"increase the deviation between "CAD"and "USD" and is where you make your money on entering when they are about to retrace (READ: Overbought - Oversold). This way your risk is reduced.
The only risk to really watch out for is risk events that strengthen (or weaken) both "CAD" and "USD" at the SAME TIME! So always check the news for the 2 "NON Hedged" currencies before palcing your trade.

Here is a list of pairs I have traded on M15 and M30 with consistent profits.
I chose pairs based on correlation periods over 6 months (120 trading days) and 12 months (240 trading days). All pairs are correlated above 80%, at least on my brokers platform. Some pairs may be missing cos my broker does not have it.

EURCAD EURUSD
EURJPY EURUSD
EURAUD EURCHF
EURAUD EURUSD
EURCHF EURJPY
GBPJPY GBPUSD
AUDCAD AUDUSD
AUDCAD AUDJPY
USDCHF CADCHF
AUDUSD AUDCAD
EURUSD EURCAD

Big Thanks to all the folks in the thread especially DL and SMJones for the Tradable Correlations Indi (Mataf.net is a distant memory now!)

I hope my input might help everyone to be a more profitable.
Regards CKP
  • #2,102
  • Feb 2, 2011 10:48am Feb 2, 2011 10:48am
Quoting Ckp
Hi DreamLiner


Ive been trading correlations for 6 months with good success and would like to share some important points that have really helped me to avoid losing trades; Time Frame and Pair Choice.

Time Frame:
In my experience M1 and M5 are just too short and consistent profits will elude you.

Ive been trading M15 and M30. If I get a signal in M15, i try to compare the same signal in M30. If M30 is still a long way off from a signal, I wait longer until I feel comfortable or until both M15 and M30 give a signal. Over time you...
Hi Ckp
How do you use tradeable correlation indicator to get correlation numbers for 240 days.That is 345600 hrs. I try to put that number on the indicator and nothing seem to happen.
  • #2,103
  • Feb 2, 2011 11:43am Feb 2, 2011 11:43am
Quoting ES Trader
Hi Ckp
How do you use tradeable correlation indicator to get correlation numbers for 240 days.That is 345600 hrs. I try to put that number on the indicator and nothing seem to happen.
ES, indi take cor period as days. so enter 240 instead of 50. no need to convert to hours
  • #2,104
  • Feb 2, 2011 11:47am Feb 2, 2011 11:47am
Quoting smjones
This is a new version of the Tradable Correlations table that has many new features. It has alerts, sorts, filters, adjustable time frames, adjustable alerting and adjustable email alerts. It has the stochastic difference and the combined spreads (in pips), not pipetts and a few more things.

I also included a doc and also a pdf (if you don't have Word) that explains it.

I hope you enjoy it. So now you need not be distracted with pairs you do not want to see, that do not meet the criteria you set.
Has anyone a similar problem with displaying of version 1.6? What's to be done?
Attached Image (click to enlarge)
Click to Enlarge Name: correlation table 1.6.jpg Size: 51 KB
  • #2,105
  • Edited 12:11pm Feb 2, 2011 11:51am | Edited 12:11pm
Quoting peje
Has anyone a similar problem with displaying of version 1.6? What's to be done?
What is the problem?

http://www.forexfactory.com/showpost...postcount=2075

Please read the PDF or doc that posted with the indicator. Thanks much, scott
  • #2,106
  • Feb 2, 2011 12:21pm Feb 2, 2011 12:21pm
I know we aren't supposed to post on this EA here but I wanted to share some results and I don't have posting privileges with Steve.

I ran the EA yesterday on a 5M time frame using an 80 entry and a 40 exit. The EA had some wins but losses outweighed them and it slowly lost equity through the day.

This morning after reading some additional posts here I changed it to a 30 minute time frame and used 80 entry and 50 exit. The EA has had 4 straight winners for around 90 pips. I'll keep testing. I have closed a couple of the trades early when they were in +20 pip territory.

Thanks to all the developers of this thread. DL SMJ Roundrock Steve and others have added greatly to the testing and I think with some tweaking this could be an overall profitable strategy. Time will tel l
Over confidence can be a moral hazard.... not to mention a financial one!
  • #2,107
  • Feb 2, 2011 12:32pm Feb 2, 2011 12:32pm
Quoting smjones
What is the problem?

http://www.forexfactory.com/showpost...postcount=2075

Please read the PDF or doc that posted with the indicator. Thanks much, scott
Thank you Scott, wrong parameter, of course, my idiocy...
  • #2,108
  • Feb 2, 2011 12:40pm Feb 2, 2011 12:40pm
Quoting peje
Thank you Scott, wrong parameter, of course, my idiocy...
Through the inputs, you can filter out pairs that don't meet the user defined Difference threshold and pairs that don't meet the user defined spread threshold and pairs that don't meet the user defined correlation threshold. You can also filter out negative correlated pair by setting the NegativeBelow threshold to something less than -100, which is mathematically impossible, and filter out positive correlations if you want by setting the PositiveAbove greater than 100. So they don't get printed to the screen. So now you only need look at the pairs that meet all of your criteria and not be bothered with the ones that do not, as you would not trade them anyway.
  • #2,109
  • Feb 2, 2011 1:00pm Feb 2, 2011 1:00pm
Quoting Mcforex
I know we aren't supposed to post on this EA here but I wanted to share some results and I don't have posting privileges with Steve.

I ran the EA yesterday on a 5M time frame using an 80 entry and a 40 exit. The EA had some wins but losses outweighed them and it slowly lost equity through the day.

This morning after reading some additional posts here I changed it to a 30 minute time frame and used 80 entry and 50 exit. The EA has had 4 straight winners for around 90 pips. I'll keep testing. I have closed a couple of the trades early when they...
On a 5 min TF, i found 80 entry and 20 exit is reasonably ok. With this setting, i earned 266 pips profit in 2 days. But i feel it would be much better to enter at 90 and exit at 10.

Also i am testing following :

5 min TF, Entry at 90, exit at 10.
15 min TF Entry at 80 and exit at 20.
30min TF Entry at 80 exit at 50.

Will post the results here as i have no privilage to post in steve thread. Also it would be a good idea to exit by pips.
Intraday only.
  • #2,110
  • Feb 2, 2011 1:11pm Feb 2, 2011 1:11pm
When i ran going home EA in 30 m TF, I get following error message in journal. What is this and why?

2011.02.02 14:05:12 Old tick GBPAUD30 1.60451/1.60482

Any one can help? Thanks.
Intraday only.
  • #2,111
  • Feb 2, 2011 1:17pm Feb 2, 2011 1:17pm
This update is EXACTLY the same as Tradable Correlations 1.6 , (functionally). I just added a text that appears on your screen when there are no pairs that meet the criteria that you set in the inputs of this indicator. You can use this one or 1.6 version they are functionally Identical.
Attached Image
Attached File(s)
File Type: mq4 Tradable Correlations 1.6b.mq4   14 KB | 1,246 downloads
  • #2,112
  • Feb 2, 2011 1:48pm Feb 2, 2011 1:48pm
Quoting navk
On a 5 min TF, i found 80 entry and 20 exit is reasonably ok. With this setting, i earned 266 pips profit in 2 days. But i feel it would be much better to enter at 90 and exit at 10.

Also i am testing following :

5 min TF, Entry at 90, exit at 10.
15 min TF Entry at 80 and exit at 20.
30min TF Entry at 80 exit at 50.

Will post the results here as i have no privilage to post in steve thread. Also it would be a good idea to exit by pips.
I agree with you on the optional exit method. Sometimes it is good to bank the pips especially when it appears it may be heading out of profit.
Over confidence can be a moral hazard.... not to mention a financial one!
  • #2,113
  • Feb 2, 2011 3:07pm Feb 2, 2011 3:07pm
Hi Scott

I am getting the following error repeatedly:

Tradable Correlations 1.6 GBPUSD,M5: ArrayCopySeries function internal error

Everything seems to be working fine but this error is being generated all the time.

I think Dryclean reported a similar error before.

Nothing urgent. Thanks is advance.
  • #2,114
  • Feb 2, 2011 3:13pm Feb 2, 2011 3:13pm
Quoting Zr-fx
Hi Scott

I am getting the following error repeatedly:

Tradable Correlations 1.6 GBPUSD,M5: ArrayCopySeries function internal error

Everything seems to be working fine but this error is being generated all the time.

I think Dryclean reported a similar error before.

Nothing urgent. Thanks is advance.
Yes I have already answered questions on this. It is because you are changing to a time frame that there is not enough history. Just change up and down to through all of the time frames with the stocdiff set at 1 and that will force history for all the different time frames on all pairs. After that there should be no more of this error. The error does not cause any functional problem. It is because the program is attempting to copy a close array for a currency that is not in your history.
  • #2,115
  • Feb 2, 2011 3:42pm Feb 2, 2011 3:42pm
Quoting smjones
Yes I have already answered questions on this. It is because you are changing to a time frame that there is not enough history. Just change up and down to through all of the time frames with the stocdiff set at 1 and that will force history for all the different time frames on all pairs. After that there should be no more of this error. The error does not cause any functional problem. It is because the program is attempting to copy a close array for a currency that is not in your history.
Thank you so much and sorry for missing your previous answer. I tried to find if you had responded to a similar question from DryClean but obviously not looked hard enough
  • #2,116
  • Feb 2, 2011 3:44pm Feb 2, 2011 3:44pm
Thanks much to everybody for their posts today, much good information and changes are coming. The 30 minute (or at least 15 minute) timeframe seems much better and I've taken several very good trades today on it.

I really do appreciate each person's input, I know we all do. Everybody testing this and coming up with variations and changes is doing nothing but helping us all. So thank you each one.
  • #2,117
  • Feb 2, 2011 4:41pm Feb 2, 2011 4:41pm
Quoting Ckp
... Trade only positively correlated pairs with a total of 3 different currencies one of which "MUST" be hedged ie: (Buy)EurCad (Sell)EurUsd. This way you are hedged for any risk event on "EURO". Risk events for "CAD" or
"USD"increase the deviation between "CAD"and "USD" and is where you make your money on entering when they are about to retrace (READ: Overbought - Oversold). This way your risk is reduced.
The only risk to really watch out for is risk events that strengthen (or weaken) both "CAD" and "USD" at the SAME TIME! So...
Thank you for sharing your useful experiences with us. I must confess when I first read through the thread it was my understanding that the buy/sell pairs (or quads as DL puts it) were selected according to the principals you mentioned (i.e. 3 different currencies one of which "MUST" be hedged. Then I noticed that this was not the case and quads consisting of 4 different currencies were considered to be eligible trades (e.g. buy GBPNZD, sell USDCHF). This did not sound logical to me and I do not trade such pairs.

If you could be so kind can you tell us if you had traded such quads and what were the results.

Kind regards
  • #2,118
  • Feb 2, 2011 4:59pm Feb 2, 2011 4:59pm
Quoting astral77
Thank you for sharing your useful experiences with us. I must confess when I first read through the thread it was my understanding that the buy/sell pairs (or quads as DL puts it) were selected according to the principals you mentioned (i.e. 3 different currencies one of which "MUST" be hedged. Then I noticed that this was not the case and quads consisting of 4 different currencies were considered to be eligible trades (e.g. buy GBPNZD, sell USDCHF). This did not sound logical to me and I do not trade such pairs.

If you could be so kind can you...
I have taken 6 quads today 2 of which are still open (1 in profit about 23 pips the other in DD about 40 pips). 4 of these 6 quads were not truly hedged pairs and so far all of the closed ones have been for 20-30 pips (2 open quads are both non hedged). I too am wondering as we test this if I want to eliminate such quads. But so far they have been some of my better trades so I will continue testing them for now.

Just my 2 cents worth.
Over confidence can be a moral hazard.... not to mention a financial one!
  • #2,119
  • Feb 2, 2011 5:31pm Feb 2, 2011 5:31pm
The question is: do the trips make the trips back home faster than the quads? If we only trade triples (EURUSD/EURJPY) and not the quads (GBPCAD/NZDUSD) have we eliminated risk or increased reward? I think the historical correlation indicator gives us our answer (thanks again SMJones!). Answer: if they are correlated 75% of the time, they are correlated 75% of the time, just like the triples.

Just my thoughts...
  • #2,120
  • Feb 2, 2011 6:09pm Feb 2, 2011 6:09pm
Quoting navk
When i ran going home EA in 30 m TF, I get following error message in journal. What is this and why?

2011.02.02 14:05:12 Old tick GBPAUD30 1.60451/1.60482

Any one can help? Thanks.
Hi Navk

' Old Ticks ' error messages have been a pain in my butt for as long as I have been trading
Previously when I had old tick errors show up in the logs I wouId simply have closed and restarted the platform
As I was only ever doing manual trading this was not much of an issue, just an annoyance

However since I have recently started using ea's this has become an issue, because it was becoming clear the ea's were being affected by these error messages when they are in the logs, and not trading

I contacted one of my brokers about this very problem and they simply fobbed me off to a forum site
You can see what they say about it here -
http://forexforums.dailyfx.com/gener...tml#post444298

I too would be very interested to hear from the more experienced ea traders / savvy MT4 techies on how ( if you can ) ' manage ' old tick errors and avoid the problems they seem to cause with our ea's
8 traders viewing now