Holmes: Can you explain the price square thing in further detail? I don't quite understand it thank you
Ok, I'll take a shot...please be kind If we're talking price change over time, sampled and measured at arbitrary but discrete time-scale interval(s), then what we have is a waveform or signal. The signal can then be analyzed with a variety of indicators and signal processing (analysis) algorithms to determine the characteristic frequency(s) and further, to identify whether the last point or sample is in-phase with one or more of frequency cycles. In general, a financial time series waveform consists of a primary frequency and some number of sub-frequencies, which may or may not be harmonics (or an octave) of the primary frequency. Also, to make things even more complicated, the primary frequency can and will change. So, in order to detect when to buy and when to sell you need to be able to: a) identify the primary frequency b) establish some certainty of its duration c) calculate the reversal times of the cycle d) understand whether and when you're in-phase with the cycle and at a point for trend reversal e) and execute on that information in a timely manner. Once all of this can be done, then it must be adaptive to changes in frequency and those changes need to be quickly identified, confirmed, and applied. Beyond all the price time-series frequency analysis, a good system also considers "other" factors and applies those to it's trading model(s). That's all I know today! Norman