How do you use correlations? | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Which is the best robot for you? That you currently use 23 replies

What brokers would you want to use if you had 100k + million + account? 6 replies

If you use a breakout model, how do you determine your TP and SL? 9 replies

Currency Correlations 2 replies

How do you use correlations?

  • #1
  • First Post: Oct 13, 2006 9:13am Oct 13, 2006 9:13am
Hi guys,
if you have a long signal on two pairs and the correlation is +0.99 you enter on one of theme,otherwise you double you position. If the correlation is -0.99 you choose one position too, otherwise the positions equal zero. What if the correlation is around 0.95?When the correlation is too strong and you have to choose one of the positions?

Candos
  • #2
  • Oct 13, 2006 10:07am Oct 13, 2006 10:07am
I don't use correlations (Heck, I don't even believe that you can trade them successfully), but I believe the idea in trading correlations is to benefit from the interest rate difference that is paid out every day. So in your examples, you would want to take position on both because then you can profit off of the interest rate difference.

Complicated subject, I'd be surprised to see if anyone was actually consistently profitably.

Hope that helps,
andersenws
"Youth is the trustee of prosperity." - Benjamin Disraeli
  • #3
  • Oct 13, 2006 10:15am Oct 13, 2006 10:15am
Quoting canados
Hi guys,
if you have a long signal on two pairs and the correlation is +0.99 you enter on one of theme,otherwise you double you position. If the correlation is -0.99 you choose one position too, otherwise the positions equal zero. What if the correlation is around 0.95?When the correlation is too strong and you have to choose one of the positions?

Candos
Here are some generally accpeted statistical guidelines where "r" represents the degree of correlation. Note that the same would apply to negative correlations.

"r" ranging from zero to about .20 may be regarded as indicating no or negligible correlation.
"r" ranging from about .20 to .40 may be regarded as indicating a low degree of correlation.
"r" ranging from about .40 to .60 may be regarded as indicating a moderate degree of correlation.
"r" ranging from about .60 to .80 may be regarded as indicating a marked degree of correlation.
"r" ranging from about .80 to 1.00 may be regarded as indicating high correlation

It's up to you to decide how much correlation is too much.
  • #4
  • Oct 13, 2006 10:26am Oct 13, 2006 10:26am
thanks..I think that using correlations is not esasy for a newbie so I'll choose pairs that are not correlated

Canados
  • #5
  • Oct 13, 2006 11:06am Oct 13, 2006 11:06am
Phil Said
Here are some generally accpeted statistical guidelines where "r" represents the degree of correlation. Note that the same would apply to negative correlations.

"r" ranging from zero to about .20 may be regarded as indicating no or negligible correlation.
"r" ranging from about .20 to .40 may be regarded as indicating a low degree of correlation.
"r" ranging from about .40 to .60 may be regarded as indicating a moderate degree of correlation.
"r" ranging from about .60 to .80 may be regarded as indicating a marked degree of correlation.
"r" ranging from about .80 to 1.00 may be regarded as indicating high correlation


Canados Replied
thanks..I think that using correlations is not esasy for a newbie so I'll choose pairs that are not correlated

************
Candos, Your exactly right in saying that. But whomever told you Trading Fx was easy, was probably a Brkr or someone looking to put others in the same pit of despair he currently occupies.

During Low Volatility periods, consider this...{Trade High Volatility if you have the Pills for it}

Long 10 Mini Lots of My Favourite Pair.
Short 5 Mini Lots of "r" ranging from about .80 to 1.00 may be regarded as indicating high correlation.

With proper control rules in place, Exit the poorer trade.

I can hear it now. "Why wouldn't I just go Long 5 Mini Lots of My Favourite Pair?"
Because if you do and it turns out sour, are you more likely to Enter an opposite trade with an additional Spread expense, or feel good about already being in the Other Trade?
  • #6
  • Oct 13, 2006 11:27am Oct 13, 2006 11:27am
I didn't mean that trading forex is easy but I meant that is better follow a system on most pairs without the complication like correlations. In fact, I'm testing a system but today there was different signals on correlated pairs.I was confused so since now I'll choose no correlated pairs
  • #7
  • Oct 13, 2006 11:30am Oct 13, 2006 11:30am
There are very few reasons to hedge in Forex. The only legitimate one that I can think of is a carry trade where you are trying to capture the interest rate differential between two pairs. That's great if you have 6 months and are happy with a pip a day. Not to mention that you still take on some directional risk while doing so.

If you want to trade Forex, pick a pair, get a system and trade it. Don't trade correlated pairs in opposite directions. That's plain retarded. Grow a spine, pick a direction and go with it. If you're wrong, stop out or reverse course. Trading correlated pairs in the opposite directions only benefits your broker since you've paid twice the spread and locked yourself out of any profit.
  • #8
  • Oct 13, 2006 11:40am Oct 13, 2006 11:40am
Quoting philmcgrew
There are very few reasons to hedge in Forex. The only legitimate one that I can think of is a carry trade where you are trying to capture the interest rate differential between two pairs. That's great if you have 6 months and are happy with a pip a day. Not to mention that you still take on some directional risk while doing so.

If you want to trade Forex, pick a pair, get a system and trade it. Don't trade correlated pairs in opposite directions. That's plain retarded. Grow a spine, pick a direction and go with it. If you're wrong, stop out or reverse course. Trading correlated pairs in the opposite directions only benefits your broker since you've paid twice the spread and locked yourself out of any profit.
Let me get this clear phil, are you suggesting I have no spine for this?
  • #9
  • Oct 13, 2006 11:43am Oct 13, 2006 11:43am
Nothing personal. I don't know you or your system. Most hedgers in Forex who traded correlated pairs against each other are spineless. That wasn't my first choice of anatomy.
  • #10
  • Oct 13, 2006 11:58am Oct 13, 2006 11:58am
Quoting philmcgrew
Nothing personal. I don't know you or your system. Most hedgers in Forex who traded correlated pairs against each other are spineless. That wasn't my first choice of anatomy.
Without Predjudice.
Your occasionally right phil. You don't know me or my system.

Perhaps because your other part of the anotomy you were thinking of are merely Virtual and therefore you didn't know how to spell it.

You tell me you consider me spineless and then when challenged you say "nothing personal". Big Weballs buddy.

I'm sure you will have a retort similar to those you have posted to TRO, Twoblink... and the list goes on... like "Oh yeah, then post your system." I've noticed you do this a few times. What's the problem phil? You don't have one?


Hey Scott, Get your eraser ready.
  • #11
  • Oct 13, 2006 12:11pm Oct 13, 2006 12:11pm
I said there are very few reasons to hedge in Forex.

If you have one, I'm happy for you.
  • #12
  • Oct 14, 2006 11:42pm Oct 14, 2006 11:42pm
Quoting philmcgrew
There are very few reasons to hedge in Forex. The only legitimate one that I can think of is a carry trade where you are trying to capture the interest rate differential between two pairs. That's great if you have 6 months and are happy with a pip a day. Not to mention that you still take on some directional risk while doing so.

If you want to trade Forex, pick a pair, get a system and trade it. Don't trade correlated pairs in opposite directions. That's plain retarded. Grow a spine, pick a direction and go with it. If you're wrong, stop out or reverse course. Trading correlated pairs in the opposite directions only benefits your broker since you've paid twice the spread and locked yourself out of any profit.
philmcgrew,

Happy with a pip a day??? You do have a lot of post on this forum but I would suggest you contribute what you know and not some bias opinion.

I have been trading for 27 years and I will post one of my trade reports for you to see. I have been in this trade since August 23, 2006. The graphic would have been too large to post here so I will put it up on a web site. My open equity is as of Friday Oct 13, 2006 is $8361. I have almost doubled my money with a "Pip a Day" hedge position. The interest isn't too bad either. Here is my site http://4xinterest.com/webpages/ratetrade.html
  • #13
  • Oct 15, 2006 7:16am Oct 15, 2006 7:16am
Quoting tortoise
philmcgrew,

Happy with a pip a day??? You do have a lot of post on this forum but I would suggest you contribute what you know and not some bias opinion.
Here's the math for everyone to see:

First, a simple formula for calculating how much money you make from interest rates.

Contract Size * [(Rate Difference / 100) / 365] = USD

Let's say you find good negative correlation between a long GBPJPY and short CHFJPY. This is currently a very popular hedging strategy.

The interest rate for GBP is 4.75%, JPY is 0.25%, and the CHF is 1.75%, therefore going long GBPJPY makes 4.5% (4.75% - 0.25%) and simultaneously going short the CHFJPY costs you 1.5% (1.75% - 0.25%). Overall, you gain the difference between the long GBPJPY and pay the short CHFJPY rate. 4.5% - 1.5% = 3.0%

Now, plug this into the forumula (we'll use 1 standard lot):

100,000 * [(3 / 100) / 365] = USD
8.21 = USD or $8.21

The current USD pip value of the GBPJPY is $8.4 for a standard lot.

8.21 / 8.4 = .98 pips.

I said it was 1 pip but it's actually slightly less. Also, these are true rates. Many brokers keep a little of the differential for themselves so it's unlikely that you would realize this ideal amount.

Math is not an opinion. It's universal and used by all countries.
  • #14
  • Oct 15, 2006 10:08am Oct 15, 2006 10:08am
LOL... Lets just put this one behind us... These seem a bit off topic and not productive... Everyone here is an experienced trader and everyone probably has more experience than me. So, How about burying the hatchet, just not in each other... LOL... What do ya say?

Scott
  • #15
  • Oct 15, 2006 10:20am Oct 15, 2006 10:20am
Quoting smjones
LOL... Lets just put this one behind us... These seem a bit off topic and not productive... Everyone here is an experienced trader and everyone probably has more experience than me. So, How about burying the hatchet, just not in each other... LOL... What do ya say?

Scott
Party Pooper....:

  • #16
  • Oct 15, 2006 10:39am Oct 15, 2006 10:39am
About how to use correlations. Lets stay on topic. thanks much, Scott
  • #17
  • Oct 16, 2006 1:36pm Oct 16, 2006 1:36pm
I use correlation all of the time. I wait for moderate seperation to occur between pairs such as eur/chf and gbp/chf for example and then in theory hedge my position by entering both trades. Or you can trade the pair eur/gbp for very similar results.

Look for iboxpips thread. He initially got me started with that. It is a rough concept in his thread... though you can learn to start from it.

Good day,
The market is my nation. Traders, my family. Hello, brothers and sisters!
  • #18
  • Oct 16, 2006 2:42pm Oct 16, 2006 2:42pm
I have been able to substantiate the existence of few correlations that are 100% reliable. However, I can state...... as pure scientific fact........ that 6-12 beers in any male subject will increase the attractiveness of ANY female subject by at least 400%.

I don't suppose that will help you in your trading efforts. But, it could save your life someday.... man. So, think before you drink.
  • #19
  • Oct 16, 2006 3:22pm Oct 16, 2006 3:22pm
Quoting philmcgrew
Here's the math for everyone to see:

First, a simple formula for calculating how much money you make from interest rates.

Contract Size * [(Rate Difference / 100) / 365] = USD

Let's say you find good negative correlation between a long GBPJPY and short CHFJPY. This is currently a very popular hedging strategy.

The interest rate for GBP is 4.75%, JPY is 0.25%, and the CHF is 1.75%, therefore going long GBPJPY makes 4.5% (4.75% - 0.25%) and simultaneously going short the CHFJPY costs you 1.5% (1.75% - 0.25%). Overall, you gain the difference between the long GBPJPY and pay the short CHFJPY rate. 4.5% - 1.5% = 3.0%

Now, plug this into the forumula (we'll use 1 standard lot):

100,000 * [(3 / 100) / 365] = USD
8.21 = USD or $8.21

The current USD pip value of the GBPJPY is $8.4 for a standard lot.

8.21 / 8.4 = .98 pips.

I said it was 1 pip but it's actually slightly less. Also, these are true rates. Many brokers keep a little of the differential for themselves so it's unlikely that you would realize this ideal amount.

Math is not an opinion. It's universal and used by all countries.
Great example. Now let's take this further. I don't trade those pairs so I hope I am not too far off when I say it cost $1,000 margin for each lot at 100:1 leverage. That would mean that 6 pairs (3 pairs each) would cost $6,000 in margin. Of course at 200:1 that margin is $3,000 and at 400:1 it is $1,500. My broker lets me set the leverage as high as 400:1. $1,500 margin and very good correlation like pholmcgew said gives me $8,500 to just sit on my position as long as I want. No stress, no fear...."so spine"

Now let's use a $10,000 account where I have 6 (2x3) of those pairs on. Long 3 GBP/JPY and short 3 CHF/JPY. Using philmagrew's math we make $8.21 a day for each set of pairs. 3 pairs would be 24.63. Now times that by 250 market days a year and that comes to $6,157.50

61% Return on $10,000!!

Now I can also add to this a trading technic that uses only a fraction of my initial position (i.e. 1 mini lot) and buy 1 mini when the market drops 100 pips and sell 1 mini when the maket goes up 100 pips. This adds and takes away from my original position. It will also add about another 10% to my return so now I will have 71% on my account.

So how have you all done in the last 12 months? Are your accounts up over 71%? Hmmmmmmmm
  • #20
  • Oct 16, 2006 3:43pm Oct 16, 2006 3:43pm
Quoting tortoise
Great example. Now let's take this further. I don't trade those pairs so I hope I am not too far off when I say it cost $1,000 margin for each lot at 100:1 leverage. That would mean that 6 pairs (3 pairs each) would cost $6,000 in margin. Of course at 200:1 that margin is $3,000 and at 400:1 it is $1,500. My broker lets me set the leverage as high as 400:1. $1,500 margin and very good correlation like pholmcgew said gives me $8,500 to just sit on my position as long as I want. No stress, no fear...."so spine"

Now let's use a $10,000 account where I have 6 (2x3) of those pairs on. Long 3 GBP/JPY and short 3 CHF/JPY. Using philmagrew's math we make $8.21 a day for each set of pairs. 3 pairs would be 24.63. Now times that by 250 market days a year and that comes to $6,157.50

61% Return on $10,000!!

Now I can also add to this a trading technic that uses only a fraction of my initial position (i.e. 1 mini lot) and buy 1 mini when the market drops 100 pips and sell 1 mini when the maket goes up 100 pips. This adds and takes away from my original position. It will also add about another 10% to my return so now I will have 71% on my account.

So how have you all done in the last 12 months? Are your accounts up over 71%? Hmmmmmmmm
Dude, that's great! Sadly, many would reject an offer for 61% a year for the dream of making more.

Trading 3 lots for 1 pip is still a pip a day though

Your return could even be higher than 61% because you haven't considered the fact that as your account grows you might add additional positions as margin becomes availalbe.

Of course, the downside is that there is no such thing as a perfect hedge and you do assume some directional risk during periods of non-correlation. You only have to scroll back to last Decemeber to see that someone blinked and the GBPJPY dropped 1100 pips while the CHFJPY hedge only covered 400 of those pips. Ouch for those who bought into the 1 for 1 correlation. There will be some nervous campers again this year as the end of the year approaches.

Thanks for discussing this like an adult. It's a fascinating strategy that a lot of people don't know about. I did not intend to imply that any single individual is without a spine. I do however maintian that if you don't know where the market is going (in the absence of a carry trade strategy) that you are better sitting out than paying double spreads to particpate in the market while not realizing any profit.
0 traders viewing now