RenTech: using artificial intelligence to morph one trading strategy into another

Quote from savagemp5:

and the system's ability to learn and recognize markets and which strategy will work or not, and thus, the morphing and adaption.
More...
I think the problem is that by the time the change in market conditions has been detected by the algo, the market may revert back to it's former behavior....thus making any move to a new strategy a losing proposition. This is the classic "whipsaw".
 
Quote from savagemp5:

I would say its a combination of multiple strategies say 1000 ? and the system's ability to learn and recognize markets and which strategy will work or not, and thus, the morphing and adaption.

btw, if anyone had the details to tell u might as well ask them to give you $10 million to your bank a/c =).
More...

There are programd that do this but I'm not sure how they escape curve-fitting even though their developers claim they do:

http://www.elitetrader.com/vb/showthread.php?s=&postid=2968698#post2968698

TSL is such an example but the price tag is just too high for individuals. At the other side of the spectrum you can find programs like PAL that are based on selecting simple patterns with high frequency of appearrance. Here selection takes the place of curve-fitting. Some claim both are negative conditions. I'm not sure... However, in all cases a lot of additional work must be done to develop an edge. The only tangible benefit when using such systems is the high throughput. Increasing the rate at which one developes and tests startegies may in itself be an edge.
 
Quote from syswizard:

I think the problem is that by the time the change in market conditions has been detected by the algo, the market may revert back to it's former behavior....thus making any move to a new strategy a losing proposition. This is the classic "whipsaw".
More...

Hint: You haven't seen the use of AI computers to analyse market data, ticks by ticks, and when I meant strategy I dont mean SMA 20/140 or MACD 10,20 for 3 months charts,

But rather the different use of at least 10-20 strategies acting on the same sets of data ticks coming in, and base in ms, the use of good strategies to costantly reap in, especially in FX markets thats quite dynamic, having different spreads around.

Its like having BAC trading at $7 on NYSE, $6.8 on ARCA, $ 7.15 on some other Unknown exchanges :)
 
IMHO: Using tick charts as the data basis for a strategy is no different than using daily bars for a longer term strategy. The only difference is the average time-in-trade.
BTW I've always been a big fan of tick bars vs. time bars as their frequency correlates to the amount of market activity.
 
Quote from syswizard:

IMHO: Using tick charts as the data basis for a strategy is no different than using daily bars for a longer term strategy. The only difference is the average time-in-trade.
BTW I've always been a big fan of tick bars vs. time bars as their frequency correlates to the amount of market activity.
More...

Heh... there's a huge huge difference, boy you miss that out. But I'll leave time to help you find out. =)

Give you a hint. Try looking at sound waves or rays... things you find in Physics. Do they resemble daily charts more or tick charts ?
 
Quote from syswizard:

IMHO: Using tick charts as the data basis for a strategy is no different than using daily bars for a longer term strategy. The only difference is the average time-in-trade.
BTW I've always been a big fan of tick bars vs. time bars as their frequency correlates to the amount of market activity.
More...

Very true.
 
Speed per tick or speed per second works far better for finding market activity and support or resistance than STD, market profile, histograms. Etc
Unfortunately market activity continues to decline
 
You must log in or register to reply here.
Back
Top