Correlation Deviation | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

MT4 MTF Correlation table and Correlation oscillator indicators 155 replies

Standard deviation indicator 3 replies

Max Deviation for MT3? 0 replies

Q: Enable maximum deviation from quoted price 5 replies

Deviation/Variability 4 replies

Attachments: Correlation Deviation
Exit Attachments

Correlation Deviation

  • #21
  • May 11, 2012 2:16pm May 11, 2012 2:16pm
May I ask if you are trading real money or are you in demo trading?
  • #22
  • May 11, 2012 2:18pm May 11, 2012 2:18pm
Right now It's kind of an odd weighted strategy...

So if lot size set to 100...
Eur/USD @ 1.29 = 129 lots
USDCHF @ .92 = 92 lots
EUR/CHF @ 1.2 = 120 lots



Makes very little sense, but with equal lot sizing returns were much less
  • #23
  • May 11, 2012 2:20pm May 11, 2012 2:20pm
I have traded baskets live before (with small lot sizes) but included JPY pairs with poor money management so my account took a hammering.

Good entries are key as you want to see the pairs out of correlation (eg. if all three in the basket are trending down), so you simple MA trending entry style seems very easy and sensible to be.

Your lot sizing isn't balanced in fact you are overexposed in strongest pairs. Interesting.
  • #24
  • May 11, 2012 2:22pm May 11, 2012 2:22pm
Quoting signorx
May I ask if you are trading real money or are you in demo trading?

For the xample of this thread, it's on a demo. I currently have it running on a real account as well, but its mixed with discretionary trading so it is much tougher to isolate and share.
  • #25
  • May 11, 2012 2:23pm May 11, 2012 2:23pm
Quoting r2997790

Your lot sizing isn't balanced in fact you are overexposed in strongest pairs. Interesting.

Correct -- its intentional. Based on testing, even though it may temporarily go against me, the frequency that it hits the 12 pip goal is much higher.
  • #26
  • May 11, 2012 2:25pm May 11, 2012 2:25pm
How long have you been testing Fortiron? i wonder if this can be properly backtested?
  • #27
  • May 11, 2012 2:30pm May 11, 2012 2:30pm
Quoting r2997790
How long have you been testing Fortiron? i wonder if this can be properly backtested?

It's very difficult to test... I usually re calibrate the MA periods once a month.

The way i tested it was somewhat ...interesting. Tested 3 main criteria...
1) Moving average periods
2) PIP Goal
3) Lot Sizing(variable or dynamic)

The moving average periods were consistently BEST in with the top value between 24-30 and the bottom value between 18-22. this was down on individual monthly tests going back to 2007.. Pain in the butt to do..

For lot sizing, it ended up being a bit of a trade off.. with dynamic lot sizing and uneven weighted like i am doing now, the 10-12 pip profit goal worked great... when i switched to equal weighting, the more optimal profit goal ended up being like 6pips... the risk of ruin wasn't significantly higher when unevenly weighting them, since i'm not SUPER high leverage, so i stuck with the higher pip goal.
  • #28
  • May 11, 2012 2:34pm May 11, 2012 2:34pm
Interesting work Fortiron. With unweighted lots did you find it took longer to close out in proft then balanced lots? Unweighted may have bigger drawdown implications too, did you see this as a big problem?

Wow, excellent job doing some 'backtesting' back to 2007. How long have you been forward testing?

Something else I just wanted to confirm, your 12pip TakeProfit I assume is the Total Basket Profit, ie. the combined profit across the 3 pairs = 12 pips then close.

I love hedging trades, safer, smaller and hopefully consistent profits as your statement showed.
  • #29
  • May 11, 2012 2:59pm May 11, 2012 2:59pm
Quoting r2997790
Interesting work Fortiron. With unweighted lots did you find it took longer to close out in proft then balanced lots? Unweighted may have bigger drawdown implications too, did you see this as a big problem?

Wow, excellent job doing some 'backtesting' back to 2007. How long have you been forward testing?

Something else I just wanted to confirm, your 12pip TakeProfit I assume is the Total Basket Profit, ie. the combined profit across the 3 pairs = 12 pips then close.

I love hedging trades, safer, smaller and hopefully consistent profits as...

It does take a little longer to close out trades, but not significantly. I find with the bigger pairs, which are weighted more heavily, the moves can be quicker, which helps.

I've been ''forward testing for only a few months.

The 12 pip take profit is much more complex- just takes the default lot multiplier, figures out the average pip cost of the open positions, and multiplies by 12 pips.
  • #30
  • May 11, 2012 3:03pm May 11, 2012 3:03pm
You don't do things the easy way

So, you work out what the average pip value is x 12 (in cash) and then when the _basket_ reaches that profit level you close the basket?

You have very creative solutions. What are you finding is your average trade duration, as rollover could also be unhelpful to a method like this.

Thank you for sharing, it will help others analyse and hopefully share thoughts too.
  • #31
  • May 11, 2012 3:19pm May 11, 2012 3:19pm
Quoting r2997790
You don't do things the easy way

So, you work out what the average pip value is x 12 (in cash) and then when the _basket_ reaches that profit level you close the basket?

You have very creative solutions. What are you finding is your average trade duration, as rollover could also be unhelpful to a method like this.

Thank you for sharing, it will help others analyse and hopefully share thoughts too.
Avg. Trade Length: 3h 0m

My solutions are 'creative' lol. Rollover is an issue, and i haven't quite figured out a way to deal with it. Aussie rolls are the worst :-/

I thought about sticking with just the EUR/CHF basket cause the rolls are much more reasonable, but we'll have to see how it works out.

Glad you are enjoying
  • #32
  • May 26, 2012 2:19pm May 26, 2012 2:19pm
This is great, hope you'll share the EA soon.
  • #33
  • Jun 6, 2012 8:52am Jun 6, 2012 8:52am
Quoting fortiron
Avg. Trade Length: 3h 0m

My solutions are 'creative' lol. Rollover is an issue, and i haven't quite figured out a way to deal with it. Aussie rolls are the worst :-/

I thought about sticking with just the EUR/CHF basket cause the rolls are much more reasonable, but we'll have to see how it works out.

Glad you are enjoying
Hello again. I was wandering are you using stop loss of any kind (one for each pair, or total basket stop loss)? Thanks.
  • #34
  • Last Post: Feb 24, 2017 2:14pm Feb 24, 2017 2:14pm
Quoting fortiron
This strategy focuses on deviations from predefined correlations. It's loosely based on triangular trading/arbitrage Its a fully automated strategy. Started on May 9th. Here's log results.... Start Balance : 50,000 Current balance: 52,284 {image} {image} {image}
Hi my brother trader, I was reading the posts of the Forex Factory members about the correlation between the pairs of currencies, your particulary caught my attention and I read it with great interest, visualizing the logs of your trades, Some questions remain unanswered:

1. I realize you are not handling the stop lost and take profit; Is voluntary or you are left in front of your computer to monitor your trades given that you open several (about ten or more) during the same day?
2. To grow your capital, how do you calculate risks and volumes at the opening of each trade ?

thank you very much
0 traders viewing now