Predicting the EUR/USD by using Currency Correlations | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Complex Correlations individual currency indicies 35 replies

Correlations- USD>JPY-DOW, EUR/JPY Follows etc 12 replies

Currency Correlations 1 reply

currency correlations 1 reply

Currency Correlations 2 replies

Attachments: Predicting the EUR/USD by using Currency Correlations
Exit Attachments

Predicting the EUR/USD by using Currency Correlations

  • #21
  • Edited 4:32pm Jan 12, 2014 4:19pm | Edited 4:32pm
Quoting PipMeUp
Brokers just don't care. The spread IS what restrict such a thing. The spread naturally comes from the lack of opportunity which reduces the liquidity. LP make money on big spread here and on a lot of small spread on the majors. They cannot make money on a lot of small spread on E/DK because they would make no money.
Yea that is what i meant, the range of the EUR/DKK is smaller than the range between the bid & ask, so if you put a long order at the ask price, you must close it on the bid price, so if marked doesnt move enough so that the BID> initial ASK = orderopenprice() ,then you end up with a loss.And i checked my broker they have 20~ pipette spread on it, while the maximum range is about 70 pipettes or so.It could be trader but is it wise to trade it?

Buy at alltimelow and sell at alltime high ? Is there enough money/liquidity in the entire market to do so? What if a 1000 pip spike appears and wipes my entire account ?

Quote
You can also build the same kind of index for each of the 7 majors. Deduce the strongest curreny and the weakest one and look at the chart of their cross ;-)
=> http://www.forexfactory.com/showthread.php?t=163158

I see but does it really matter which one i build it.The objective would be to pick 1, where i picked the EUR/USD because its popular & low spread, and try to find autocorrelations in the composed pair made up of those other currencies which then can predict the EUR/USD in real time.

In case the EUR USD would be too random, or less information carrying in the candlesticks about the future price, but still a global event has been happening which will effect the EUR/USD in some way, in this case maybe the other currencies can show better the future price of the euro, then the euro itself.

Autoregression in my correlator indicator can very well predict the EUR/USD, whereas in the EUR USD it could be harder to detect ?
What about mean-reversion tehniques, a few carefully picked moving averages can smoothen my indicator, and could also show us some hidden information about the EUR/USD ?

------------

BTW Pipmeup you are giving me an excellent idea

The way how hannover did i could construct an indicator which shows the EURO and the DOLLAR's strength separately, by using all available EUR & USD containing couples.

This could also show the direction of the EUR/USD not only the correlations with itself, what a nice though, thanks!
"There's a sucker born every minute" - P.T. Barnum
  • #22
  • Jan 13, 2014 4:55am Jan 13, 2014 4:55am
Just for fun:
Prediction:
EURUSD +,USDCAD+ ->EURCAD +

The Euro seems to be coming up from resistance in a range: if this holds, the
USDCAD has resumed its trend, and the implication is that the EURCAD will as well.

I have a trade going on the EURCAD just above what appears to be resistance. A pending buy there should take and go into profit before the US open. If not, remove and watch.

Proximus is right : the u shape pattern I have described does appear to be of different origin. Lot's of good ideas on the thread, guys, keep it up!
  • #23
  • Jan 13, 2014 7:27am Jan 13, 2014 7:27am
Quoting Rap Skallion
Just for fun: Prediction: EURUSD +,USDCAD+ ->EURCAD + The Euro seems to be coming up from resistance in a range: if this holds, the USDCAD has resumed its trend, and the implication is that the EURCAD will as well. I have a trade going on the EURCAD just above what appears to be resistance. A pending buy there should take and go into profit before the US open. If not, remove and watch. Proximus is right : the u shape pattern I have described does appear to be of different origin. Lot's of good ideas on the thread, guys, keep it up!
Now , at 8am EST, the pattern has not held: EURUSD is down, USDCAD is down and EURCAD is down ( The relation ship held, but all three signs went the other way.

Now I would remove the pending entry, and wait for the us open. If the boys in New York agree with the boys in Europe(does not always happen) , look to re-install a pending trade, could be a buy, could be a sell, depending on
the way the cookie crumbles. Fairly high probability the three will move together. (Just which way is the question).

Meanwhile I am still making money on the GBPJPY some 50,60 pips, plus yesterday. Pardon my precision.
  • #24
  • Jan 13, 2014 8:46am Jan 13, 2014 8:46am
Quoting Rap Skallion
Just for fun: Prediction: EURUSD +,USDCAD+ ->EURCAD + The Euro seems to be coming up from resistance in a range: if this holds, the USDCAD has resumed its trend, and the implication is that the EURCAD will as well. I have a trade going on the EURCAD just above what appears to be resistance. A pending buy there should take and go into profit before the US open. If not, remove and watch. Proximus is right : the u shape pattern I have described does appear to be of different origin. Lot's of good ideas on the thread, guys, keep it up!
You're forecasting EUR/CAD based on EUR/USD and USD/CAD??? You are kidding, aren't you?
That's a cross. EUR/CAD = EUR/USD * USD/CAD at any point of time.
No greed. No fear. Just maths.
  • #25
  • Jan 13, 2014 8:50am Jan 13, 2014 8:50am
In my enthusiasm I neglected to restrict myself to the EURUSD. I will correct this from now on.

There are three implications I currently watch that imply a rise in the EUR\USD:
AUDUSD+EURAUD+87.4 A AND B 98.2
USDCAD-EURCAD+88.2 A AND B 91.0
USDJPY-EURJPY+87.4% A AND B 98.7




Let me interpret one of those:
If the AUDUSD rises, AND the EURAUD drops the EURUSD should rise (other things being equal)
Correlation between the EURUSD & the EURAUD is currently 87.4, so when the one goes down, the other one usually does too.
There is a very high correlation between the AUDUSD & EURAUD, so that the signal generated by this triplet is unusual: relatively rare< and a break in the usual pattern. Breaks in patterns can be a source of profits.


Reverse the signs:
if the AUDUSD drops and the EURAUD rises the EURUSD should fall (other things being equal).


Simply put, these three pairs rise and fall together frequently.


I am still working on this stuff. All that thinking hurts my brain. This could all be in error, so
play with it and see if you can make something of it. Let me know if you can.
  • #26
  • Jan 13, 2014 10:10am Jan 13, 2014 10:10am
Quoting PipMeUp
{quote} You're forecasting EUR/CAD based on EUR/USD and USD/CAD??? You are kidding, aren't you? That's a cross. EUR/CAD = EUR/USD * USD/CAD at any point of time.
Precisely: Crosses should, but don't always, hold true. The market is not a logical machine.

I would be guilty of belaboring the obvious if it weren't that so many are oblivious to this relationship.

This quality of the Forex doesn't hold true with stocks and bonds, and people coming from that area of endeavor understandably are not looking for it, since no such relationships are widespread between stocks or bonds.

I gave a screen shot a while back in this thread that shows it does not always hold true.

You begin to understand, I think...
  • #27
  • Jan 13, 2014 10:47am Jan 13, 2014 10:47am
Quoting Rap Skallion
{quote} Precisely: Crosses should, but don't always, hold true. The market is not a logical machine.
I'm concerned about what you mean by this statement. The E/C will always, without fail, always, exactly equal E/$*$/C. So, would you please explain what you mean by your statement?
  • #28
  • Jan 13, 2014 11:37am Jan 13, 2014 11:37am
Quoting GEfx
{quote} I'm concerned about what you mean by this statement. The E/C will always, without fail, always, exactly equal E/$*$/C. So, would you please explain what you mean by your statement?
It should and doesn't. Isn't that fascinating?
one possible explanation:

You have a group of people interested in the EUR/USD: suppose there are more buyers than sellers: the price goes up.

You have another group of people interested in the USD/CAD : again more buyers than sellers: the price goes up.

Now you have a third group of people interested in the EUR/CAD: But the buyers of the first two just are not looking at the third, (Lots of people don't trade across the board). In this group of people there are more sellers than buyers: the price goes down.

Now, if the same group of people were interested in all three, this would never happen.

In a truly efficient market, everyone would know what everyone else was doing, and (one assumes reasonable people) this would never happen.

But greedy and fearful people (and the guy who just got a message from Mars) are not rational.
And knowledge is not equal.
So it does happen. (sometimes to my chagrin).

The trick is figuring out how to take advantage of this irrational, inefficient wrinkle.

That's why I say "other things being equal."

This is not the only thing that can be unequal, but isn't that enough?

Now I hope you have something you can play with, and possibly profit from.

Good luck. Any genuine refutations gladly accepted and acknowledged.
  • #29
  • Jan 13, 2014 11:46am Jan 13, 2014 11:46am
Actually what i meant and i think Pipmeup thinked of this too is to separately measure the strenght of the EURO and the USD, by using all avaliable pairs which contain either EUR or USD

So the formula for USD is: 1/xx1USD * USDxx2 * USDxx3 * 1/xx4USD*... ,where the xx are other pairs like NZD,AUD,JPY, all except the EUR, and the pegged currencies like DKK,CHF and so on

And the formula for the EURO is: 1/xx1EUR * EURxx2 * EURxx3 * 1/xx4EUR*... ,where the xx are other pairs like NZD,AUD,JPY, all except the USD, and the pegged currencies like DKK,CHF and so on

This way i have an indicator which measures the EURO strength, and the USD strentgh separately, then i have to put them on an indicator window and compare them, if the EURO > USD = EURUSD bullish if the EURO<USD = EURUSD bearish.

I think i explained it clearly
"There's a sucker born every minute" - P.T. Barnum
  • #30
  • Jan 13, 2014 12:13pm Jan 13, 2014 12:13pm
Quoting Rap Skallion
{quote} It should and doesn't. Isn't that fascinating? one possible explanation: You have a group of people interested in the EUR/USD: suppose there are more buyers than sellers: the price goes up. You have another group of people interested in the USD/CAD : again more buyers than sellers: the price goes up. Now you have a third group of people interested in the EUR/CAD: But the buyers of the first two just are not looking at the third, (Lots of people don't trade across the board). In this group of people there are more sellers than buyers: the...
I trade all the majors and many of the exotics, and have never seen, for example, the E/J not exactly equal EUR/$*$/JPY. If someone would please show me an example, I'd love to see it, aside from an example of a minor delay in prices from your broker that equals some fraction of a pip, or an increase in spread due to some goofy broker problem/issue.
  • #31
  • Edited 1:49pm Jan 13, 2014 12:25pm | Edited 1:49pm
Quoting GEfx
I trade all the majors and many of the exotics, and have never seen, for example, the E/J not exactly equal EUR/$*$/JPY. If someone would please show me an example, I'd love to see it, aside from an example of a minor delay in prices from your broker that equals some fraction of a pip, or an increase in spread due to some goofy broker problem/issue.
I only trade the E/J, G/J. and E/$ and totally agree that the E/J is a function of the E/$, and $/JPY. I'm not sure what you mean by being "Exactly Equal" though? What i see is that at certain times the E/$ is causing the E/J to move, other times it's the $/JPY driving, and even others it's demand from the crosses.

Now in terms of correlation the E/J, and G/J are typically very strongly correlated and make for a good pair to execute cross currency hedging on
  • #32
  • Jan 13, 2014 12:59pm Jan 13, 2014 12:59pm
Quoting yazzie
{quote} I only trade the E/J, G/J. and E/$ and totally agree that the E/J is a function of the E/$, and $/USD. I'm not sure what you mean by being "Exactly Equal" though? What i see is that at certain times the E/$ is causing the E/J to move, other times it's the $/USD driving, and even others it's demand from the crosses. Now in terms of correlation the E/J, and G/J are typically very strongly correlated and make for a good pair to execute cross currency hedging on
Hi yazzie, my point is that the E/J is only derived from the E/$ and the $/Y. To assume that you can catch a trade where there is any material advantage (due to some external factor or force other than the three currencies in question), is just wrong. The question as to what is driving the E/J price (something dollar related, yen related, or both) is another issue.
  • #33
  • Jan 13, 2014 1:24pm Jan 13, 2014 1:24pm
On re-examining my charts, I think I see some confusion. Good! that makes the relationship a lot tighter. So if I see the first two of the pairs in solid trends, the third pair will be also. A simple matter of logic. opening a trade on the third would then depend almost directly on what the first two are doing, and closing the trade on the third may be indicated when the logic no longer holds.

So if some catastrophic news causes a plunge in the EURUSD and the USDJPY rises I will be able to trade the EURJPY perhaps avoiding the widened spread? Lots of people trade that pair two, so the spread is probably widen there as well.

So this can only be used for confirmation, not prediction. (That is what I originally thought).

The scenario I had presented is possible, but unseen. I take the word of someone who has traded for 5 yrs.

Now as to whether Proximus' idea of using currency indexes and correlations somehow to predict , more or less the EURUSD. Can you calculate the odds? Can you tell reliably when they are in your favor?

This bears thought and re-reading. Thanks for the clarification!
  • #34
  • Jan 13, 2014 3:18pm Jan 13, 2014 3:18pm
Quoting Rap Skallion
On re-examining my charts, I think I see some confusion. Good! that makes the relationship a lot tighter. So if I see the first two of the pairs in solid trends, the third pair will be also. A simple matter of logic. opening a trade on the third would then depend almost directly on what the first two are doing, and closing the trade on the third may be indicated when the logic no longer holds. So if some catastrophic news causes a plunge in the EURUSD and the USDJPY rises I will be able to trade the EURJPY perhaps avoiding the widened spread? Lots...
I guess I am not following very well. Could you briefly outline the scenario again? It seems to me that unseen really just means, it never happens.
  • #35
  • Jan 13, 2014 3:28pm Jan 13, 2014 3:28pm
I think you understand me. I have never seen it happen, at any rate. Some others have cautioned me that the arrangement does not always work. (Jees! some five years of trading and I am just figuring this stuff out!)
  • #36
  • Jan 13, 2014 4:03pm Jan 13, 2014 4:03pm
Quoting Rap Skallion
I think you understand me. I have never seen it happen, at any rate. Some others have cautioned me that the arrangement does not always work. (Jees! some five years of trading and I am just figuring this stuff out!)
I don't understand, but good luck to you, and good trading!
  • #37
  • Jan 15, 2014 6:05pm Jan 15, 2014 6:05pm
I'm back with the crosses:

If AUDUSD goes down and EURAUD goes down. Then EURUSD should godown, right? The money should flow from the EURO thru the AUD to the USD.
also if the USDCAD goes up and the EURCAD goes down the same thing should happen, right? The money should flow from the EURO thru the CAD to the USD.
Also if the USDJPY goes up and the EURJPYgoes down , the same thing:


Now these are seven different pairs, none of which are pegged on the the other. Does this mean that they all act in concert? that there is no chance that while the first two crosses are occuring that the USDJPY cannot go down and the EURJPY up all at the same time.

I think from time to time this occurs. Which way then does the EURUSD go? The money flows from the EURO through the CAD and the AUD to the USD and back from the USD thru the JPY to the EURO? (my head hurts).

I just won't trade that particular configuration.

There is , I think a point to all this, which I will try to explain later..
  • #38
  • Jan 15, 2014 7:09pm Jan 15, 2014 7:09pm
Quoting Rap Skallion
I'm back with the crosses: If AUDUSD goes down and EURAUD goes down. Then EURUSD should godown, right? The money should flow from the EURO thru the AUD to the USD.
You have the math correct, but your question does not make sense. If the EUR/AUD is going down, it's because the Eur/$ and Aud/$ are going down. I think it's important to see that no matter how much you buy or sell the EUR/AUD, there is no material impact on the Eur/$ or the Aud/$. Perhaps you didn't mean what I implied from your question.
  • #39
  • Jan 16, 2014 2:11am Jan 16, 2014 2:11am
Quoting GEfx
If the EUR/AUD is going down, it's because the Eur/$ and Aud/$ are going down.
E/U and A/U can both go up while E/A goes down as long as A/U is going up faster. E/U and A/U can go down and E/A up as long as A/U falls faster.
No greed. No fear. Just maths.
  • #40
  • Jan 16, 2014 2:32am Jan 16, 2014 2:32am
Quoting PipMeUp
{quote} E/U and A/U can both go up while E/A goes down as long as A/U is going up faster. E/U and A/U can go down and E/A up as long as A/U falls faster.
Of course this is true, but that is not the point I am making. You can buy or sell all the E/A you want, and that will have no material impact on the E/$ or the A/$. The value of the E/A only changes when the values of the E/$ and or A/$ changes.
0 traders viewing now