Partial quote from cd23:
04-21-08 01:06 PM
This is brief comment that will, hopefully clarify a few things. I will do the details later.
Have read through your most recent post (and numerous others) and it is clear that the volatility seam must be the trend-supporting pace. Originally thought that there was an IF-1/2 logic involved with delineating the volatility subsets but believe now that isn't the case. Rather it is a HH/HL, LH/LL, insides and outsides thingy which are operative. This construct takes care of the directionality aspect. That the OB should be a restart makes common sense but where I'm for the moment running the hamster wheel is the mechanism by which the other components are derived.
Given that the distributions of pace and overlap are Gaussian (normal), there must be some sort of critical point (? the midpoint of the distribution) which serves to discriminate between continuation and reversion, where the latter means a falling volatility. Also to make this paradigm workable there must be some truncation of the wings of the Gaussians.
It would help if I could have a bit more info about the acronyms and abbreviations. I think the hi/low comp refers to the comparison of the high and low of the developing bar to that of the immediately prior bar but what is OR?
The extremes of high pace and minimal overlap or low pace and maximal overlap are easy to envisage. The conditions for change must have to do with BO from the VDU/DU states and polarity changes. The behaviour of laterals can be readily extrapolated from what's known already. I don't see any component which is comparable to the very helpful Volume/Volatility anomalies (high volume-low volatilty and low volume-high volatility) both of which are powerful predictors of change.
I appreciate that there is more than a trifling component of "pay as you go learning" (as in me frying a Betz cell or two) with your challenge and if I'm coming up short then I'll check back later with new info and more ignorant queries.
lj