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

guys, when trying new pair with this method, they must be cointegrated to reach the mean 0. If not, then this exercise is meaningless. If you look at the cointegration link i posted, this test will tell you if its cointegrated or not.

I finally made it work(it was a problem with the rpath afterall).I also noticed that the maximum bars arbomat can analyze is 2500. I am like a kid playing with its new toy here but i dont know how exactly this toy should be used.I dont fully agree with what i read about arbomat not repainting. Yes, the lines dont repaint, but the mean and the standard deviations constantly change with each new bar.The good news is that this "change" is so small that it practically doesnt exist.So ,although not theoritically true, we could say that arbomat doesnt repaint.
What i cant figure out is how we can use this tool most profitably.In my opinion it is useless to create regressions with more than 2 pairs,as they wont be fundamentally backed up.For example,there is no reason to bet that the +0.39EURUSD-0.12USDCHF+100AUDUSD-15.5USDJPY is going back to its mean after making an extreme because there is nothing backing that up.On the other hand,betting that 100AUDUSD-99NZDUSD is going back to its mean after making an extreme, will probably be a highly profitable bet since these 2 pairs are very highly correlated and this correlation is fundamentally backed up.
So, we should find the most hilghly correlated pairs,add them in arbomat's code and trade based on their correlations.
Concerning the selection of timeframe,i am totally lost.I dont know whether it would be better if we stayed on longer timeframes or on shorter ones.I ll have to think about this.
One last note: I dont think that placing you trades based on MT charts is going to be profitable.MT shows neither the mean of the spread nor the deviation.It just shows the spread.For example,lets say that two pairs are perfectly correlated.What does this mean?It means that their gap should be fixed(this could be 100 pips,200 pips or even 100000 pips).So,you go on MT4 and you see a 200 pip gap between these 2 pairs.This doesnt tell you anything.You cant bet whether this gap wil go to 0 or not.You need to know the mean and the deviation of the gap.
If the mean of this gap is for example 400,it is more probable that the gap will widen rather than go to 0.

will I just start playing with this system. I am thinking of entering trades in both direction and deal with each basket separately. since it is hedge I think it should work fine
lets try and see what could happen.!!

I demoed it but just the opposite what you did. When it closed to 0 , I opened two pairs basket. When its value increased, the profits increased

Can you please post a screenshot? Thanks!!

I opened the basket at -14.5, now it is -27.5 and profit is already > $50.

I've also taken this trade 5min ago!

Good Luck!

Not understand this

Normally you must Sell EURUSD and Buy GBPUSD , or I understand something wrong

Icemaan

For either of you does your R chart look like this for 1 min eurusd / gbpusd?

Maybe when the spread is positive you must Buy EURUSD and Sell GBPUSD and when it is negative you must Sell GBPUSD and Buy EURUSD or viceversa, I am testing this right now...


this is how it looks to me right now with 2000 and 5000 bars back

This is my chart

Is that 1 min tf?

Edit: NEVERMIND the second thing i talked about!

Just closed $150 Profit

Yes this is 1 minute Chart
But what I must do now

1. Buy EURUSD or Sell EURUSD
2. When the Spread is negativ Arbomat show me the correct Values and I only must change when Spread is positve?

Thanks for Help

icemaan

hmm, I noticed your R chart only based on the last few hrs of data. Mine is from 5/17 9:37 to current. This changes the deviation levels, but amazingly both seem equally valid. I'm up around $20 on this basket entered right before posting the r chart

I always make the sign opposite, not sure if this is right but i think it is.

So in this situation since eu is 100, you would sell 1 lot.

With this Method you win when the Spread go away from Zero , but i think the Goal is that the Spread returns to Zero , or I`m totally missunderstand this Trading Concept

icemaan

I think your way is better than mine since I have no idea what happened. When do you close your trades?

IMO its highest percentage to place trades near mean cause it has to deviate in either direction. After sufficient data analysis has been performed, trades can be placed at certain deviation levels, anticipating a move back to the mean. This has been done in the stock market for years. Lookup mean reversion trading in google.

right before I posted.