You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Who has any ideas on what principle, or rules, to close orders in profit. For simplicity, let it be EURUSD/USDCHF.
Just compare the forward sums for each closing option and it will be clear.
Please give me an example.
If I understand correctly, the comparison has to be done all the time. Still, I do not understand how to catch the best point.
If you don't have a problem with maths, see: albert-shiryaev.pdf
Thanks for the link.
I am not very good at maths, but I have a question - have Shiryaev and his comrades-in-arms become millionaires?)
Please give me an example.
If I understand correctly, the comparison has to be done all the time. Still, I don't understand how to catch the best point.
Thanks for the link.
I'm not very good at maths, but I have a question - have Shiryaev and his companions become millionaires?)
EUR-CHF correlation coefficient, daily data, 2009-2015, floating window of 400 observations.
Pair trading in forex is impossible - the correlation is "floating" from -1 to +1 for all instruments, it is impossible to build a market-neutral portfolio.
But you can read the methodology for building, opening and closing here - https://ru.wikipedia.org/wiki/%D0%9F%D0%B0%D1%80%D0%BD%D1%8B%D0%B9_%D1%82%D1%80%D0%B5%D0%B9%D0%B4%D0%B8%D0%BD%D0%B3
Made a model based on COINTEGRATION a few years ago. Even discussed it on the Fours forum.
The model was very decent. Especially attractive was the fact that trading decisions were made on a stationary (or very close to it) time series.
The model did not work as the major part of profit was up to 5 pips 4-digit, while with my broker at the time it was usually 2 pips, very often 6, and sometimes even 20 pips....
Today it is quite possible to apply on EUN accounts. By all means a treasure trove for pipers.
But my interests have changed and I don't do cointegration anymore.
PS.
The model itself is a couple of dozen lines on R.
Pair trading in forex is impossible - correlation is "floating" from -1 to +1 for all instruments, it is impossible to build a market-neutral portfolio.
Correlation has to float, which is partly what we earn on.
For USDCHF, one hour history is available only from 2015.04.23. Compared to euro, counted from period separators, (not the best prices), everything works.
On EURUSD, on the undeflected trend, also works, but the algorithm is different there. Calculated from period separators on H4
Pair trading in forex is impossible - correlation is "floating" from -1 to +1 for all instruments, it is impossible to build a market-neutral portfolio.
Correlation must float, which is partly what we earn on.
For USDCHF, one hour history is available only from 2015.04.23. Compared to euro, counted from period separators, (not the best prices), everything works.
On EURUSD, on the undeflected trend, also works, but the algorithm is different there. I calculated it using period separators on H4.
The correlation should not be floating.
It's very easy and simple to check it - EURUSD*USD=EURCHF. If these two symbols are highly correlated by the coefficient -1, the Major should move in a narrow horizontal channel. Trading is simple - open a position when the price reaches the border of the channel, and close it when it reaches the opposite border.
You can NOT see it on the chart. In chunks, yes, but you cannot determine such a segment in advance.