- Search Energy EXCH
- mbkennel replied Feb 27, 2011
There's a reason for that: you never know for sure if the rules give an edge in forward testing. Things "going bad" is a good sign that the edge has disappeared. Ideally I'd want some "minimal sufficient statistic" (only in analogy to the technical ...
Best trading systems on forexfactory
- mbkennel replied Feb 27, 2011
I realize I was wrong in the above. It's similar, but not very similar. Cutting the worst performing according to phantoms bases P&L on the open point of the phantoms and not the individual trade's opening point. I don't think this has been ...
Phantom 6 System
- mbkennel replied Feb 27, 2011
Isn't that a typical time for turnarounds? Can you motivate? And what would be the criterion? If you are in margin danger then the trading session could really screw you well before the once-per-day time, unless this is for a different reason. ...
Phantom 6 System
- mbkennel replied Feb 26, 2011
This is the bug that I work around in LibOrderReliable (hint hint). The SELECT_BY_TICKET internal database may not be the "the same" as SELECT_BY_POS,MODE_TRADES Look to see what I do in O_R_CheckForHistory() in LibOrderReliable4.mq4
Brijon Concept of Threes auto trading robot by Steve Hopwood
- mbkennel replied Feb 26, 2011
Cutting the worst performing is very similar to the cut loser, which sometimes is OK but often isn't. Only difference is "when?".
Phantom 6 System
- mbkennel replied Feb 18, 2011
Hypothetically, if you traded fully balanced (i.e. adjusting a lower lotsisze for EURxxx than for USDxxx, and a little higher than AUDxxx) one pair of each should result in no profit, just loss from transaction costs. If you're not fully hedged, ...
Phantom 6 System
- mbkennel replied Feb 18, 2011
No, it is a modified version of the one I posted a while ago. (Phantom6_01_x_CL_TH_PS_BAL2) Generally, leverage=100:1, mmtu=0.65, pp=0.5, pipstep=10, cutloser=1.5 which works great on B EURCHF,EURJPY,CHFJPY; S GBPCHF,GBPUSD,USDCHF and absolutely ...
Phantom 6 System
- mbkennel replied Feb 18, 2011
I actually have a test EA with about 10 different pair sets built-in (all synthetically hedged and running with my fully balanced lotsize mod). An EA with the original set seems to be chugging along with good profits. One of the others went into 15 ...
Phantom 6 System
- mbkennel replied Feb 17, 2011
How to implement JCS's margin protector — This is my idea of a 'margin protector' per JCS, which disrupts the baskets the least. We will maintain two global variables per basket, both reset to zero upon closure/flat position: number of trades ...
Phantom 6 System
- mbkennel replied Feb 16, 2011
Because I wouldn't want a trade at 7 or 7:15 to start without some momentum developing.
Phantom 6 System
- mbkennel replied Feb 15, 2011
Yes, logistic regression is an option. what you want. The underlying model for logistic regression is that the logarithm of the odds, i.e. log(prob(true)/prob(false) is a linear function of the inputs. It is the simplest generalization of linear ...
Relation between binary and continuous variables? (In MatLaB)
- mbkennel replied Feb 15, 2011
Sometimes though the profit gets taken on an extreme fluctuation and resetting there might mean you can get the snapback as well. I'm not sure about it either. If I had to choose one point to reset, it would be 5 AM London time. (GMT in winter and ...
Phantom 6 System
- mbkennel replied Feb 14, 2011
If the Big Banks and futures industry scoop the business by offering a more regulated and honest trading (do you think Goldman would try to screw AQR or Renaissance this way?) I'm 100% for it. The one really important piece of a lawsuit is subpoena ...
Class Action Lawsuit Against FXCM
- mbkennel replied Feb 12, 2011
Charvo is right. With lots of leverage and trades and small profit percent it is easy to make something which seemingly looks better than it really is. Always think: what are the hidden risks? We need to make it safer and maintain the profitability. ...
Phantom 6 System
- mbkennel replied Feb 10, 2011
Wow. I wrote the baseline of that a long time ago. I have no idea if it is worthwhile but it is an unusual combination from my weird brain.
Optimized Trend Trading
- mbkennel replied Feb 10, 2011
Actually it's not so hard. I wrote libGMT.h which looks at the Windows time (assuming you set your local computer time and time-zone correctly, which most people do) and figures it all out (getting local time of day in NYC/London/Tokyo correcting ...
Beastie: a Nanningbob trend and counter-trend auto-trader by Steve Hopwood
- mbkennel replied Feb 9, 2011
Remove the EA. Clear all global variables. Put the EA on again .
Phantom 6 System
- mbkennel replied Feb 9, 2011
Hi Scoobs. If, optimally, you wanted to define a "day", what times would those be? e.g. start time in GMT or London local and close time? Perhaps one could start 8AM tokyo to 5pm NY as a "day" and ignore even the remaining few hours? It should be ...
Beastie: a Nanningbob trend and counter-trend auto-trader by Steve Hopwood
- mbkennel replied Feb 8, 2011
Ask Scooby what they do. (he trades for hsbc). They rely on properly researched algorithms to inform exceptionally skilled individual traders. They spent lots of money (millions hiring programmers & PhD's) on fully automated systems, but even at ...
Self-help mentality