System to hedge every major dollar pair for profit / low DD | Page 20 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Any tools to know what 1 pip is for every currency pair? 1 reply

Price info indicator needed - one pair's price x another pair 5 replies

Multi-pair trading or Few-Pair Trading. [poll] 15 replies

Trading 9 pair every day 143 replies

How nearly EVERY pair gap downward?! 8 replies

Attachments: System to hedge every major dollar pair for profit / low DD
Exit Attachments

System to hedge every major dollar pair for profit / low DD

  • #381
  • May 26, 2012 11:04am May 26, 2012 11:04am
will watch carefully how this voodoo stuff will unravel .....cheers
wo-yoy! wo-yoy! wo-yoy! wo-yoi! wo-yoy-yoy-yoy!
  • #382
  • May 26, 2012 11:15am May 26, 2012 11:15am
Attached Image

puzuzu shall guide us
Quoting pemully
will watch carefully how this voodoo stuff will unravel .....cheers
Skype: heliosphan187
  • #383
  • May 26, 2012 11:25am May 26, 2012 11:25am
Is there any chance we can modify Arbomat so that it can be used with all pairs?
I had an idea.If this can be done,we will be able to trade pegged currencies(currencies with fixed rates).For example,we know that EURDKK is fixed around 7.46 with a 2.25% band(in reality the central bank of Denmark has never let the currency go above or below the 1% range).So what we can do is trade the spread of USDDKK with EURUSD and of other similar fixed currencies almost risk-free.

And the other thing i have been thinking is whether we can trade such fixed currencies alone.What i mean is whether we can make R give us the mean and the standard deviations of these fixed currencies.Well,i guess we know the mean (since they are fixed),but it would be very useful if we could know the standard deviations as well.Actually i think we dont even need the R.This could be done by a simple indicator inside MT4,right?
  • #384
  • May 26, 2012 11:40am May 26, 2012 11:40am
Quoting Indrek
Here is a perfect tool for playing with correlations: http://www.forexticket.co.uk/en/tools/01-01-correlation
Nice!!!
  • #385
  • May 26, 2012 11:54am May 26, 2012 11:54am
Quoting Kruspe
I suppose it depends on strategy, I mean -> you can trade it from extrem to mean or from mean to extrem (trend following).. or from top to bottom (+ is sell and - is buy) and from bottom to top (+ is buy and - is sell).

or am I wrong DB ?
DB,

I think we should post this in post 1 so that newbies know how to enter the trades.
Follow me to the Promised Land
  • #386
  • Edited 1:04pm May 26, 2012 12:11pm | Edited 1:04pm
Quoting Dirtybrown
Attachment
puzuzu shall guide us

Yup. I am sure Puzuzu may be of help....

Alternatively, maybe we could all educate ourselves as to how all this stuff actually works...

Radical concept, I know, but certainly worth a shot.

Maybe even, as a really last resort, read my thread on all this (Old Dogs Taming of the Beast).

Best regards

Old Dog
  • #387
  • May 26, 2012 3:54pm May 26, 2012 3:54pm
Quoting mwduke
It seems to be profitable to do the following:

If Arb-O-Mat line is currently BELOW the mean, BUY pair that is a POSITIVE value (in upper left corner of R graph); SELL the pair that is NEGATIVE (if any are; there does not necessarily need to be a negative pair, especially if only inputting two pairs).

If Arb-O-Mat line is currently ABOVE the mean, SELL pair that is a POSITIVE value; BUY the pair that is NEGATIVE value.


It makes the most sense to enter this trade (either above or below the mean) once the Arb-O-Mat line is far away from the mean...
Thanks
  • #388
  • May 26, 2012 6:08pm May 26, 2012 6:08pm
Just finished reading Old Dog's Taming of the Beast thread. It's only 3 pages long and quite full of info.

One point Old Dog made that stuck with me was the idea of constructing these baskets from uncorrelated currency pairs. Key post He states that the best baskets contain cointegrated pairs that are uncorrelated.

The currently favored combos in this thread (EURUSD/GBPUSD and NZDUSD/AUDUSD) seem to be ruled out by Old Dog's statement. It's not that they won't work for a while but that they'll potentially blow up suddenly. Old Dog's approach is to come up with a basket and weights that will be stable for days to weeks.
You are in a maze of twisty little passages, all alike.
  • #389
  • May 26, 2012 7:47pm May 26, 2012 7:47pm
Quoting stevegee58
Just finished reading Old Dog's Taming of the Beast thread. It's only 3 pages long and quite full of info.

One point Old Dog made that stuck with me was the idea of constructing these baskets from uncorrelated currency pairs. Key post He states that the best baskets contain cointegrated pairs that are uncorrelated.

The currently favored combos in this thread (EURUSD/GBPUSD and NZDUSD/AUDUSD)...
Any suggestions for pairs that fit your description? If not, how do you think we should go about determining which pairs are both cointegrated AND uncorrelated (and are one of the USD pairs available in the Arb-O-Mat)?
  • #390
  • May 27, 2012 12:21am May 27, 2012 12:21am
Quoting Dirtybrown
Ok must be a local problem them, I'm loading arbomat to another broker and will try it there.

For anyone wanting to trigger the R charts from arbomat over the weekend run this script.

Thanks.
I tried this, but no luck. I've also tried copying the code manually, both versions by 7bit, but still nothing happens. Anyone with a similar problem? Any ideas how to fix it?
  • #391
  • May 27, 2012 4:17am May 27, 2012 4:17am
Hi friends
Can anyone who understands the method should kindly post the rules of the method the best time frame to trade the method

The Buy and sell rules

Thanks
Much luv
If you don't control your emotion it will control you.
  • #392
  • Edited 5:37am May 27, 2012 4:40am | Edited 5:37am
We have too much theoretical noise around... Could someone give an example of a well cointegrated basket, please? I know that USDCAD and NZDUSD are a very good combination but what about a bigger basket? What 6 or 7 pairs should be included?

I think, I am not the only person here who has no degree in math and who's native language is not English...
  • #393
  • May 27, 2012 4:55am May 27, 2012 4:55am
im just trying to understand but could a currency strength meter be used as an aid to the overall trend direction of these currencys as a help for entry maybe its a silly question?

im still learning and this is a little above me im not sure yet if i can get this system to run yet on my mt4 ill wait till monday its a liitle complicated but im learning
  • #394
  • May 27, 2012 7:01am May 27, 2012 7:01am
Here you have something that you can test the behavior even when the market is closed.
Attached File(s)
File Type: mq4 send tick.mq4   1 KB | 318 downloads
  • #395
  • May 27, 2012 7:12am May 27, 2012 7:12am
Hey Guys, I love the work and all the effort you are pouring into this project and if I can, I would like to help by demo testing and posting charts.

I have a question: I'm not a programmer and I have read the whole thread but when you download R 2.15.0 where do you install that program? I assume to C drive but do I simply create a new file folder for it? Could you be a little more clear for us non-techie types.

Thanks for your help.
  • #396
  • May 27, 2012 10:26am May 27, 2012 10:26am
Quoting FXTopGun
Hey Guys, I love the work and all the effort you are pouring into this project and if I can, I would like to help by demo testing and posting charts.

I have a question: I'm not a programmer and I have read the whole thread but when you download R 2.15.0 where do you install that program? I assume to C drive but do I simply create a new file folder for it? Could you be a little more clear for us non-techie types.

Thanks for your help.
Hey FXTopGun.

You can install it anywhere according to what I know. Just remember to set your path to where you install the R software. Also remember to set the path to the 32bit version. I think the folder is i368 (or something similar)

Let me know if having any troubles. I struggled initially, but could figure it out eventually. Oh and I also have no programming skills so we are in the same boat.

Cool Runnings...
  • #397
  • May 27, 2012 10:28am May 27, 2012 10:28am
Attached Image (click to enlarge)
Click to Enlarge Name: 17.jpg Size: 181 KB

Here we are working on something similar, yet there are lots of different sizes, therefore, is the middle curve just like the one in this thread, I think it is a good way.http://www.forexfactory.com/images/icons/icon10.gif
  • #398
  • May 27, 2012 12:01pm May 27, 2012 12:01pm
I have also read Old Dog's thread and that was one reason why I started to test hedging all usd pairs without usdchf and nzdusd. This seems to produce a better price pattern.

Quoting stevegee58
Just finished reading Old Dog's Taming of the Beast thread. It's only 3 pages long and quite full of info.

One point Old Dog made that stuck with me was the idea of constructing these baskets from uncorrelated currency pairs. Key post He states that the best baskets contain cointegrated pairs that are uncorrelated.

The currently favored combos in this thread (EURUSD/GBPUSD and NZDUSD/AUDUSD)...
Skype: heliosphan187
  • #399
  • May 27, 2012 12:52pm May 27, 2012 12:52pm
Quoting Dirtybrown
I have also read Old Dog's thread and that was one reason why I started to test hedging all usd pairs without usdchf and nzdusd. This seems to produce a better price pattern.
Do you suggest to include EURUSD, GBPUSD, AUDUSD, USDCAD and USDJPY in the basket? And trading it with intercept set to "true"?
  • #400
  • May 27, 2012 12:58pm May 27, 2012 12:58pm
I just posted a reply to 7bit's old thread thinking I was posting to this one.
Anyway here it is again.

Quote
Originally Posted by Dirtybrown
Well I am now starting my own experimentation in R being totally new to using R or any true statistical package besides excel. I will follow this example for finding cointegrated pairs http://quanttrader.info/public/testForCoint.html

I realized after reading that web page how close arbomat is to this; the one thing lacking in arbomat is the cointegration test. Arbomat calculates the spread correctly but never verifies cointegration. It looks like it'd be easy to add by just using the adf.test function and checking it against a threshold:

Inserted Code
ht <- adf.test(sprd, alternative="stationary", k=0)

cat("ADF p-value is", ht$p.value, "n")

if (ht$p.value < 0.05) {
    cat("The spread is likely mean-revertingn")
} else {
    cat("The spread is not mean-reverting.n")
}
Unfortunately I'm in the middle of rebuilding the carburetor in my tractor right now LOL.
You are in a maze of twisty little passages, all alike.
1 trader viewing now