DislikedSarcasm does not help here. It is ok, to bring up your opinions, but please do respect each other.
Thank you.Ignored
I am trying it on Demo account. But struck at the loss of equivalent to the total spread. I am curious how do you guys make profit when the spreads for 3 pairs add together is quite sizable...
Any enlightened guys care to answer?
Obviously, there is some sort of market anomaly here or some other factor at play which is beyond my comprehension
.G'day to you all.
- #143
- Feb 8, 2011 3:54pm Feb 8, 2011 3:54pm
- | Commercial User | Joined Jul 2009 | 5,375 Posts
Viewer Discretion Advised: Shall we shag now or should we shag later? :-)
You're welcome. The change to your trade sizes will do nothing but give you a complete hedge, give or take some units, and will never show a profit. Unless it's from the little exposure you'll have because you not using fractional units and you'll being paying spread and interest. No complete hedge like you will be doing is positive interest so if you let it run it will cost you everyday. At least it won't be a total waste of time cause you'll probably learn from it and gain some experience on how pair interact and are related. Simple things like...
I did experience that exact scenario.
Never thought you'd be able to turn me around that fast.
Thank you for your lessons
As for the Stochastic filter... the EA hasn't taken a trade yet, which isn't good enough for me, so I will stop posting in this thread if the EA hasn't taken a trade in over a week.
Hello the-game,
would you be kind enough to share your ea ? i would like to try it by myself.
Thanks a lot
would you be kind enough to share your ea ? i would like to try it by myself.
Thanks a lot
I must say, I will be taking interest in your posts from now on.
I did experience that exact scenario.
Never thought you'd be able to turn me around that fast.
Thank you for your lessons
As for the Stochastic filter... the EA hasn't taken a trade yet, which isn't good enough for me, so I will stop posting in this thread if the EA hasn't taken a trade in over a week.
Let me comment a little (and summarize)
In the first post of this thread, the poster thought this method is risk free. After all, using 3 pairs we have perfectly hedged each currency involved. However, maybe cash4u2 didnt realize, but this method is NOT risk free if the lot size for all 3 positions are the same. As previous posters lonestar and the-game have pointed out, if all 3 lot sizes are the same, the result is a net long or short position; basically we have just '(re)created' a pair with a very high spread!
lonestar's example calculation:
buy 10000 gbp/usd = long 10000 gbp/ short 16050 usd
sell 10000 gbp/chf = short 10000 gbp/ long 15352 chf (16050 usd)
buy 10000 usd/chf = long 10000 usd/ short 9559 chf (10000 usd)
gbp: 10000 long and short cancels each other out. No exposure.
usd: 16050 short and 10000 long leaves 6050 short usd exposure
chf: 15352 long and 9559 short leaves 5793 long (which is approx 6050 long usd)
I can think of 2 ways to remedy this.
First, the most obvious is just to resize your lots such that there is the same amount of long and short for each currency. This should be completely risk free, as the net movement for each currency is zero. However, then how would you profit? (read up more on triangular arbitrage if you're interested)
Now the second method makes more sense to me. We just have to resize each lot such that each lot makes the same amount of money per pip.
For example,
long 1.05 lots of gbp/usd: 1 pip = 1.05 usd
long 1 lot of usd/chf: 1 pip = 1.05 usd
short 1 lot of gbp/chf: 1 pip = 1.05 usd
There is still a net exposure (a net short of about 69850 usd, short of about 58630 chf, and long of 5000 gbp)
HOWEVER, the movement of each pair will be such that assuming perfect correlation, there will be only net movement of 1 pair.
(eg when gbp/usd moves up 1 pip, gbp/chf moves up 1 pip, usd/chf moves down 1 pip. Net movement is 1 pip up.)
We know that correlations between pairs fluctuate and change over time. We just need to wait until the correlation relationships among the 3 pairs involved change till we get a nice net profit and close the trade.
This seems to be how cash4u2 is making profit. If that is the case, I do not see any advantage of this strategy over traditional correlated pairs trading strategies. I am no expert in correlation trading and stat arb, so I could be horribly wrong.
This correlation aspect has been touched on by the-game and the other posters talking about the stoch EA.
As of now, looking into the mechanics of this strategy I am rather skeptical... Looking at cash4u2's posted profit results, he has about just as many winning trades as losing trades, with no clear stop loss. Maybe cash4u2 or someone else has something new to add?
I dont mean any offense, this method just seemed interesting and I thought I'd offer my 2 cents
In the first post of this thread, the poster thought this method is risk free. After all, using 3 pairs we have perfectly hedged each currency involved. However, maybe cash4u2 didnt realize, but this method is NOT risk free if the lot size for all 3 positions are the same. As previous posters lonestar and the-game have pointed out, if all 3 lot sizes are the same, the result is a net long or short position; basically we have just '(re)created' a pair with a very high spread!
lonestar's example calculation:
buy 10000 gbp/usd = long 10000 gbp/ short 16050 usd
sell 10000 gbp/chf = short 10000 gbp/ long 15352 chf (16050 usd)
buy 10000 usd/chf = long 10000 usd/ short 9559 chf (10000 usd)
gbp: 10000 long and short cancels each other out. No exposure.
usd: 16050 short and 10000 long leaves 6050 short usd exposure
chf: 15352 long and 9559 short leaves 5793 long (which is approx 6050 long usd)
So we are left with appox 6050 short usd/5793 long chf (6050 in usd terms) which is about 6050 units of short usd/chf exposure. Everything is approximate cause rates were changing as I was typing.
So just place the short u/c trade and save yourself some spread and margin.
I can think of 2 ways to remedy this.
First, the most obvious is just to resize your lots such that there is the same amount of long and short for each currency. This should be completely risk free, as the net movement for each currency is zero. However, then how would you profit? (read up more on triangular arbitrage if you're interested)
Now the second method makes more sense to me. We just have to resize each lot such that each lot makes the same amount of money per pip.
For example,
long 1.05 lots of gbp/usd: 1 pip = 1.05 usd
long 1 lot of usd/chf: 1 pip = 1.05 usd
short 1 lot of gbp/chf: 1 pip = 1.05 usd
There is still a net exposure (a net short of about 69850 usd, short of about 58630 chf, and long of 5000 gbp)
HOWEVER, the movement of each pair will be such that assuming perfect correlation, there will be only net movement of 1 pair.
(eg when gbp/usd moves up 1 pip, gbp/chf moves up 1 pip, usd/chf moves down 1 pip. Net movement is 1 pip up.)
We know that correlations between pairs fluctuate and change over time. We just need to wait until the correlation relationships among the 3 pairs involved change till we get a nice net profit and close the trade.
This seems to be how cash4u2 is making profit. If that is the case, I do not see any advantage of this strategy over traditional correlated pairs trading strategies. I am no expert in correlation trading and stat arb, so I could be horribly wrong.
This correlation aspect has been touched on by the-game and the other posters talking about the stoch EA.
As of now, looking into the mechanics of this strategy I am rather skeptical... Looking at cash4u2's posted profit results, he has about just as many winning trades as losing trades, with no clear stop loss. Maybe cash4u2 or someone else has something new to add?
I dont mean any offense, this method just seemed interesting and I thought I'd offer my 2 cents