I'm not sure if anyone else was curious about this: Quote from bwolinsky: "The holy grail is the third derivative over various ma period lengths. The change tells you where it has moved to, the change in the change tells you the concavity, and the third derivative's change in the change in the change shows how the convexity from one point to any other rotates around the other. Identifying trend, spotting bottoms and tops, and the intermediate oscillations of the market as they happen. " I coded it and it looked just like the RSI or money flow. You could adjust the time scale and again, just like any momentum indicator. I knew it wouldn't be that easy to find another edge...
The various optimized lengths are used to mark pricing levels for support and resistance onto a chart then. As much time as it takes to follow, I've integrated a variable called TSI and it has the effect of smoothing over non-extremum returns. I hope your calculations of these movements including an identification of trend with these price points marked on each of the charts. I would love to see your code by the way.
I'm waiting on the fact that you do not have anything but a tax lien living with an estranged wife in her house which she bought long before she knew you well enough to take you in?
Took me a decade and my conclusion is down below. Believe it or not, price has very special peculiarities. Consider price the human race and the instruments the humans that make this race. Consider yourself the trader, a psychologist. Study your subjects and you will begin to understand why price acts the way it does but most important where it gets irritated, complacent, confused or euphoric. Accept that price has free will and it has different options at any given time, but very limited options at very special specific crossroads. Crazy A
A simpler way to do this and be able to visualize as well as have an ATS is to do the "pin wheel". As most know it is not possible to take derivatives of non continuous functions. An equivalent stat has to be substituted. I can't remember when we made it popular but is was probably before ET was put on line. (certainly before blowinsky) Use three MLR lines that are double each other (10, 20 and 40 were used long ago). Display them on price. If you compressed time and made a camtasia of the right portion of a chart that is almost stationary on a display screen, you could see or imagine how the three MLR's form a pattern like the flow of fish flocks or flocks of birds. (there is now a math for this as well.) In detail, you just observe (or for an ATS calculate and get signals) the angular velocity among the three lines; it translates into a set of leading indicators of price that almost anyone can use for timing trades on any fractal. Note how the add/delete function keeps the signals from lagging (this is as compared to the use of EMA's or AMA's which you may have used to circumvent the Calculus dilemma) It is a neat solution for the problem that differential Calculus cannot be applied to the non continuous functions of the market. The holy grail is close to taking 3 to 6 times the daily ATR; Blowinsky is not even close because his stuff is so lagging. As I remember blowinsky finds taking the full offer of the markets "unbelievable" so far in his life.
Thanks for that..... I'm going to spend some time trying to understand it. It's true, there is a lag in using numerical methods to get the third derivative. However, with any derivative, it's very noisy the higher it is. I tried 1, 2 and 4 day spans for calculating the slope. The 1 day looked just like a noisy ROC indicator. Simpler tools like momentum divergence or higher pivots as the second impulse wave begins, seem to show more potential. Classical technical analysis..... who would have guessed?