Since when do you need rationality for a trend? Complete rationality does not exist in humans, anyway, except in the mentally deficient. See Taleb's book. --opm8
I agree, but what i don't understand is why everyone want's to test random data. The market is human-driven. Rational and/or irrational behaviour is influencing the evolution of the ticks. You don't have that element in random data. So testing on random data is useless to me. What am i missing?
Q http://classes.entom.wsu.edu/526/Exogenous.htm "When a time series is found to be non-stationary it may be necessary to model the exogenous effects before we can forecast future population densities. There are four major situations that give rise to non-stationary time series -- transients, trends, discontinuities, and multiple equilibria." UQ Q www.brettsteenbarger.com/Short-Term Trading2.doc " A major conclusion was that the distribution of price changes through the day is nonstationary, making it hazardous to employ the same buying and selling parameters through the day. " UQ Q http://epub.wu-wien.ac.at/dyn/virlib/wp/showentry?ID=epub-wu-01_1e1 "In this paper we investigate the potential of the analysis of noisy non-stationary time series by quantizing it into streams of discrete symbols and applying finite-memory symbolic predictors." UQ
What character in the actual market behavior is rational and/or irrational? Also, in what way does it influence it? Please read back to hanketsurfer's post about his subjective view of chaos and order...
mr. subliminal, i know i am a little late to this thread, but i will ALWAYS trade a random system with introduced trends. that's a no brainer. i've done sufficient money management tests to know that the only thing you need is a little trendiness, not too high commissions and spread and you'll win. Therefore i am thinking an indicator which displays trendiness could be very helpful to screen markets. Does anyone know such an indicator? I know "trendiness" by itself would show a curve, but then you'd have to trade this curve as well
1. Random system? Random system on "pseudo"-random data? 2. Introduced trends = Not random. Isn't it??? 3. What's trendiness? You might know but others might not.