I'm trying to develop a system that is long something until the price goes parabolic. I have had no luck with that Parabolic SAR, is there any indicator designed to measure price acceleration?
Keltner bands is a indicator I use. 20 day EMA with 2.25 ATR. Could use a rule that if the open is above the top band exit or if “x” number of opens are above top band. The bars should stay within those settings like 98 percent of the time, if the low gets past it things are going real parabolic.
There are some issues with using these indicators in my dataset (a lot of stocks go parabolic quite early in the data so there is not much time to calculate previous results for the indicators). What I was thinking was to convert the stock price over a certain period of time into an angle, the closer that angle is to 90 degrees, the more parabolic. Does anyone know how to do this conversion in excel?
The Parabolic SAR is quite flexible, and has 3 parameters to it. Have you understood and worked on each one? A price|time combination becomes a single point on a graph of time. An angle is a ratio, not a single point. If what you wish is a function which 'increases at an increasing rate' (something which could be termed 'parabolic'), then you want an exponential function -- as in FVn = PV*(1+r)ⁿ
I solved the problem with the formula =DEGREES(ATAN(LINEST(J4:J5))) J4 is the Price 1 and J5 if the subsequent price in which I want to see the angle. Its working quite well in terms of showing me when a parabolic move is happening
The difference between J4 and J5 would be the "Acceleration Factor" in the Parabolic SAR -- this might save you some time and screen clutter: https://stockcharts.com/school/doku.php?id=chart_school:technical_indicators:parabolic_sar (...although, the tweaking and tuning and fiddling can certainly be fun...)
I like to find things as they are entering "parabolic stage." I then get into the position and continue to ride it in the same direction. Often a parabolic move that is detected, does not end for several months, on the upside and downside. It often continues and that is the easiest way to make money...In my opinion.
%% Since you are wanting to limit profits[get out early]; momentum[price speedometer] gets you out early + that's why i dont use it.