Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Strategies & Market Trends : E-Mini Pit -- Ignore unavailable to you. Want to Upgrade?

To: Nemer who wrote (4942)	8/1/2002 1:55:11 PM
From: SE	Read Replies (3) \| Respond to of 11288

Beans, Toss all the crap and throw the dart. Sheesh, 36 screens, various markets, differing time frames, yada yada, yabo doo. Holy Cow batman. You do need to explain futher the ROC of ROC....as I don't see how QC can do that. This ole dummy is lost. -Scoot

To: Nemer who wrote (4942)	8/1/2002 9:47:55 PM
From: Dan Duchardt	Read Replies (2) \| Respond to of 11288

I hope you will allow an interjection here from an old physics professor who can't resist the urge to call for a consistency of units, recalling as I do that inconsistencies in this area were so detrimental to the understanding of those grappling with the mathematical descriptions of their observed universe. The units of a rate of change of a rate of change (an acceleration) would not be (feet per second) per (feet per second). Rather, the units of acceleration are (feet per second) per second. Your stated ratio fps/fps is a dimensionless quantity representing a comparison of two speeds, which is something human beings are quite adept at doing visually, even when the events are not simultaneous. For example, in a horse race where all the horses are running at about (not exactly) the same speed, almost anybody can answer the question of which horse is running faster at any moment. And if a person saw a horse trotting by, and later another one galloping by, they could still make the comparison correctly. They are also generally able to see when a single moving object is speeding up or slowing down. An experienced observer could do a fairly good job picking which of two galloping horses was faster, even if they came by separately. This is the only sort of comparison where the units fps/fps have any meaning; it is a comparison of two speeds, not a measure of acceleration. People generally do a very poor job visually comparing accelerations. They would be hard pressed to tell when a slower horse is accelerating more than a faster horse at some point in time, and would most likely surmise the faster horse had the greater acceleration even though that is often not the case. This is probably why your detailed calculations at the track gave you some edge over others who depended only on their visual perception. To go one level further and talk about visual comparisons of "jerk", well that is simply not something people can do without instrumentation. Just to be complete, the units of jerk are ((feet per second) per second) per second. If indeed you find detailed evaluation of subtle changes in rates of change applicable to the price movement of stocks or indices, my hat is off to you. There are some very sophisticated mathematical techniques for predicting the future state of a time evolving entity based on its current state, including some number of time derivatives (e.g. Kalman filters and extensions)) that have proved very useful in applications such as inertial guidance. From what I have been able to find, their applications to market behavior have met with limited success, no doubt because it is far more difficult to understand the relationship between the current state of the market and any future state resulting from market "noise" and the vast set of market forces that are constantly emerging and disappearing than it is to understand the effects of Newton's laws and generally well understood forces on a moving object. Were it not so, the mathematicians would have all the money. <ggg> In a broader sense, lots of people use change of ROC without calculating the rate of that change or calling it such. Everyone who plots a trend line and observes it breaking, or observes price "going parabolic" is detecting a change (though unmeasured) in the rate of change. By the time sufficient data is collected to average out the noise and measure the rate at which the slope changed, it's usually pretty obvious the change has occurred. I doubt the detailed knowledge is of any great value by then, but maybe you know something I don't know. The real trick is not in detecting that ROC has changed. Rather, it is the combined determination of the new ROC and prediction of its duration that is the key to any trading decision. I'd be very interested in any calculation that could have told me within 3 minutes that the NDX low at 10:26 this morning was just the beginning of a minor bounce, while the one at 2:07 yesterday was the beginning of a 20 point rally.