concurrent uncorrelated positions | Page 4 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Reasonably uncorrelated basket trading 6 replies

Trend basket by linear regression of uncorrelated pairs 12 replies

Should I be concerned about uncorrelated markets? 7 replies

uncorrelated pairs for risk control and focus 3 replies

Uncorrelated pairs 1 reply

Attachments: concurrent uncorrelated positions
Exit Attachments

concurrent uncorrelated positions

  • #61
  • May 10, 2014 7:18am May 10, 2014 7:18am
Quoting hanover
{quote} A pleasant 'problem' for a trader to have. {quote} It certainly beats Douglas Adams' answer of "42".
42? Oh my! I thought it was 41! Recalculate! Recalculate!
  • #62
  • May 12, 2014 5:50pm May 12, 2014 5:50pm
Quoting Rap Skallion
{quote} Ordered the book. Obviously with the limited number of pairs on offer there is bound to be some correlation but at least you can go with pairs that do not share any currency. Any correlation less than one is preferable to putting extra trades in the one currency. Thank you.
Received the book. Worth every penny. so much is revealed in a clear light. Thank you . "The Trading Game" by Ryan Jones, (John Wiley & Sons co. 1999) should be required reading for every one looking at this sentence. His critiques of common mistakes committed, nay propagated by ff writers are Thorough and reasoned.

No it will not tell you when to enter a trade, etc. but it will tell you what size to trade and what to do with your money when you get it.
Had I read this before I spent a dime, I could have saved thousands!

Mr Hanover, once again you have forced me to rethink. Such a pleasure!
  • #63
  • May 14, 2014 4:43am May 14, 2014 4:43am
Quoting shellsnail
{quote} Haha this will work very well if there's positive serial correlation... hmm but what if the serial correlation is negative wouldn't you shoot yourself in the foot? just wondering, or am I not making sense. {image}
Stupid me! We don't need the trades to be serially correlated. We need the equity curve to be. Let E be the expectancy of the system. The equity Q[t] before MM will evolve as Q[t+1] = Q[t] + E + e. This is an AR(1) process with a positive coefficient.
No greed. No fear. Just maths.
  • #64
  • May 14, 2014 5:23am May 14, 2014 5:23am
Quoting PipMeUp
{quote} Stupid me! We don't need the trades to be serially correlated. We need the equity curve to be. Let E be the expectancy of the system. The equity Q[t] before MM will evolve as Q[t+1] = Q[t] + E + e. This is an AR(1) process with a positive coefficient.
Hmm... The equity curve is not serially correlated, it is the error terms in the equity curve that are serially correlated.

E(Trade) = expected value of each trade which is definitely less than the size of your wins... Hence the error terms will be positive when you have a win and negative otherwise. If you trade a fixed RR system then you can only have two different sizes of error... The error terms are a function of your trades, so yes you still need the outcomes of your trades to be serially correlated. (i.e. a winner will tend to follow another winner)

The equity curve you stated is actually a random walk with drift (unit root process) if there is no serial correlation - i think...?
Build good relationships with others.
  • #65
  • May 14, 2014 7:11am May 14, 2014 7:11am
E is the expectancy of the system. It is a constant. The return of one trade is pnl = E + e. e is zero mean (since it's the difference with E).
Q[t+1] = Q[t] + E + e has got E as a deterministic constant trend. Upward iff the system is a winner.
In the formula from Wikipedia c is E, epsilon is e and phi is 1.
=> So you are right.
Attached File(s)
File Type: xlsx doesntWorkMM.xlsx   29 KB | 299 downloads
No greed. No fear. Just maths.
  • #66
  • Last Post: Nov 25, 2018 7:02am Nov 25, 2018 7:02am
[quote=permanentjaun;175167]{quote} If you're using the same strategy for both pairs then if they're correlated pairs they will very often give you the same entry times. Even if they give you different entry times, they will tend to end trades very similar, both winners or both losers. So it doesn't make sense to trade two pairs that are very correlated. You won't get many more trades. It'll be the same as trading one pair, but with twice the number of lots than your standard trade. So its beneficial to trade uncorrelated pairs, because if they don't have the same trading pattern then while one could be in a bad trading pattern, the other does not follow the pattern and could be trading well.

Very well explained , thanks
But many questions i have in mind
Lets asume that i risk 3% of capital per trade
I see a long signal in nzdusd , nzdjpy , and a short signal in eurnzd
Then what should i do ? Should i just check that which pair is much better for buying nzd and just trade that one pair with 3% ?
Should i trade 1% on each ?
In my exprience in this sutiation , one pair will run better than others but we dont know which one
0 traders viewing now