duard, Mandelbrot's models of fat tails have "applicability" to describing the distribution of sizes of price fluctuations, but they do not provide or suggest any method for profitably predicting whether price will go up or down. Mandelbrot, furthermore, has never even attempted to model chaos in prices.
mahras2, You say that stephencrowley isn't selling anything. But he said, in a previous posting, that he isn't giving away any freebies. If you are both correct, then the only logical conclusion is that he has absolutely nothing whatsoever to offer.
I don't care if mandelbrot says prices are unpredictable or not, I doubt he has studied super high frequency data very much. In fact, in this Article in wired he suggested that acquiring high frequency data would cost "20 million dollars" http://www.wired.com/wired/archive/12.08/view.html?pg=2 Of course, $20 million is not enough: Even if doctoral students are cheap, proprietary data are not. But with that starting amount and wise leadership, such a commission would quickly draw contributions from others, magnifying its impact. So, if he thinks high frequency data is this expensive it probably shows he hasn't messed with it much. His ideas about multi-fractality, scaling and self-similiarity, fat tails, etc, is what I'm referring to when I say he is right. Prices do not follow a log-normal distribution. And, you my friend, are a fucking idiot, I'm not selling anything, and if you offered me $100,000 grand on the spot for my software I wouldn't give it to you. Soon as I get my trading signals built around this and get it going realtime I will definitely start a journal, and then you can come there and claim that I'm lying etc, etc, it'll make for great entertainment.
I'm offering inspirition to those who want to do so. And I also get a kick out of ruffling your feathers.
I understand and agree. What Mandelbrot did as a scientist and mathematician was to look at financial markets and try to understand mathematically if there was a way to describe the action. Mandelbrot is not a trader nor is he an economist. My point was after careful study of much of his work as a trader that is what I found to be true. This was my interpretation of his findings. There are still ongoing discussions about random walk and likely always will be. Many people have traded "Volatility" profitably for a long time. In fact the most profitable traders are probably the one's who do. I'm sorry if my comment was taken as Mandelbrot's as that was not the intent. Florida's weather pattern this summer, volatility clusters in the SIF's, sunspots for that matter all cluster in a non-random but "chaotic" fashion, yes?
I never said you were a liar or a seller. I implied that such an explanation would fit the evidence you provided us, but like I said, I will keep an open mind if you can provide some further evidence. I'm glad you are finally admitting that you haven't actually traded your method. This provides further confirmation of my assessment that you do not understand the scientific literature or the technical jargon you are using, and that you have no valid prediction method. Sorry to disappoint you, but you did not succeed in ruffling my feathers. Your use of profanity and ad hominem personal attacks only detracts from your own credibility.
Wtf? I never claimed to have traded my model yet. I like to fully and completely understand (to the extent that this type of thing can be understood) what I'm doing before I put money on the table. Also, I thoroughly enjoy using profanity and my 'credibility' doesn't affect how I make money, so I really don't care.
If I recall, scientists generally accept the existence of chaos in weather and in sunspots, but that chaos in the volatility clustering of price changes has never been demonstrated. Mandelbrot has written extensively about volatility clustering in price changes, but he has never attempted to demonstrate that it is due to chaos.