Anyone else spread trading the 7 major .../JPY basket pairs too? Want to discuss more on developing a system to automate the spread. I am just using ATR indicator (long the pairs with lowest ATR and short the highes) and have achieved this result:
Frank, notice that he's opening simultaneous positions on different pairs; i.e., using a "basket." You can't just pick the losing pair from the basket and take its loss as the DD. You need to look at the results from the entire basket. I'm definitely interested in this for a project we're involved with, so please keep posting.
I understand and agree but, basket or not those losses are way out of line and indicate some type of fundimental flaw in the logic of the "system" and a very high risk of ruin. Now, if he's able to get a handle on those losses he might be on to something. Baskets with P/L like that never pan out in the long run. Sure you can curve fit it on some particularly tasty in-sample data only to get crushed by a perfect storm of out-of-sample data. I could be wrong and the OP has found the Holy Grail of FX baskets but odds are, and as the old expression says, "There's nothing new under the sun". Keep us all posted.
I agree with the fact that he needs to provide more data, but not with the fact that a loss in one of the components in the basket means that it is not a good system. If I was trading a basket, I would like to have the different positions go in different P/L directions for diversification. From the limited data that he has provided, it seem that this is what's happening.
I'd expect loss in any basket but not of the size noted in this case. Granted it's a tiny sampling (which should give even greater pause before declaring it "Superb") but the P/L and individual component drawdown is out of the norm for basket trading on any instrument. A basket with 41,948 in winning components and 23,665 in losing components does not inspire my confidence (in spite of the alleged 18293 profit) and I'm very risk tolerant. OP: Back test on 12 months data and exclude 9/1 to date and let's see what you come up with.
No need to backtest...just trade with real money like real traders. would like to discuss more with traders with basket spread/ hedging trading experience...on rebalancing/scale in-scale out optimisation for FX baskets. PM me if you want a more private analysis or find out more on the concepts behind EA,expert advisors/algorithms for FX basket trades. Anyone willing to share more info on the concept of inverse or drawdown hedging techniques?
Your posted results reflect the short-swap trade -- all of your wins are the result of long the major carry; EURJPY, GBPJPY, NZDJPY, USDJPY. You're hedging with short CHF as it's nearly neutral on swap. The ATR is highest in the pairs you are LONG. I ran a matrix long the vol and short the neutral yen swap [CHFJPY], and it produced 132% at 33:1 leverage, but the CHF hedge was 5:1 of the book.