Very interesting discussion you are having here guys. There is no doubt in my mind that patterns do exist. I use simpler OHLC patterns because I think that they curve-fit the least. I wonder though what will happen to them after the introduction of pattern mining programs like price action lab and the fact that their use is getting widespread. Will these patterns stop working? In his blog the pattern mining guru gives examples of patterns with an average life of 10 years. Do you think that use of these programs will decrease or increase the average life of patterns?
Thanks for the picture. I had done something in Excel, the picture helps me confirm I was on the right track. Slight digression -- does anyone know if there is a field of study about paths? (only requesting a name, I'll research on my own) Often, I'm not concerned about the change amount between 2 points, but the path getting from the first to the second point. The steps look like a way I could code the price movement as a path.
Path integral by Feynman, Least action principle in quantum electrodynamics...by Feynman...etc, etc, etc, by ... Feynman.
No convincing argement yet in favor of randomness in the markets. I still think the market is completely deterministic, we just do not know the initial conditions and the model. Lack of knowledge not = randomness
The first person of the singular is best in this particular case... Would you like to see the Lagrangian or the Hamiltonian for the market, or perhaps both?.:eek:
Here you are trying to impose your own definition, not the one accepted by most everybody else. Random in practice simply means not predictable. If someone can't predict something because he has insufficient knowledge it is random to that person. Consider the original use of the the random walk -- understanding the Brownian motion of small particles like pollen grains in water. Physicists realized that this is due to molecules of water bouncing against the particles. According to you this isn't random because in principle we could follow the trajectory of every water molecule to find out what will happen next to the particle. In real life, where most of us live, we have no hope of knowing the position and momentum of every water molecule (even in a classical world) so Brownian motion is treated as a random process. Indeed the statistics of the motion match a three dimensional random walk model with high accuracy. Most people would see the tossing of dice as a random process, even though in principal you could follow the physics of the muscle movement during the toss and so forth to determine how the dice will land. A casino that tries to predict how dice will land will waste a lot of time. A casino that treats dice as random and sets the win/loss amounts appropriately will make a lot of money. There are so many actors in the stock and commodity markets that prediction of future prices is very hard to do with odds any better than chance. As a practical matter price behavior is very random.
Thank you for the reference, I'll try to study it (but I think it's going to take quite some time ...)
Good luck in your journey...and happy random thoughts after you get there!. Now that the discussion has ended for all practical purposes. I will respectfully leave all of you with your PhDs, and hire whomever I need with any PhDs or programming degrees for $50 an hour... They even bring me cappuccinos...