Hey, I'm thinking of using a basic frequency table to determine support / resistance levels. The bulk of the work would involve grouping the results into classes that represent a support / resistance's range, while disregarding any frequencies that are too small to be statistically significant. I'm aware that this will only give me "obvious" horizontal S/R levels. Anyone had success with this approach ?
They work just as well as human-defined S/R levels. So, 50/50. Though I think you can mine a few more win % out of it.
Yeah I'm mostly using them as part of the various "anomalies" I'd be including in the final decision process. By themselves they might not mean much (as you said 50/50), but I'm hoping to find relationships that could increase their significance.
I would chime in and offer some thought provoking suggestions but, with names like "diapercoder" and "nooby_mcnoob", I'd rather not waste my time with children.
Someone must be really idiotic to believe that entering a discussion, leaving nothing of value and going on with their life is a good use of their time. It reminds me of my daughter wasting her life watching useless cartoons - adults can truly act like children even with a generic name like HurricaneUS Clearly your attitude is not childish. Allow me to continue walking on your lawn and enjoying doing so.
That is a weird drive by comment! But for me, my name represents that I never want to be so confident that I know everything.