fractals, the universe...let's get into it

Originally posted by chasinfla
I could barely pronounce 'Benoit Mandelbrodt,' much less understand him.

Would someone care to, again, please define 'chaos' for the purposes of this discussion? Thanks.
More...

And as my previous posting shows, I did not remember the spelling of his name either .....

BM was a pioneer: many have expanded his work. Chaos has a very technical defintion and can be approached from a couple of viewpoints. Essentially the idea is that a small change in the input of a function results in large changes in the output such that the output may not even be predicted. Classic examples are the logistic function which exhibits a phenomena called period doubling and also chaos. Period doubling phenomena are interesting and tie into the notion of self similarity and chaotic behaviour. There is a lot more to this. It is very interesting. However it takes quite a bit of experience IMHO to make use of the ideas in the market.
 
if you really want to get a grip on this stuff, you will need to read a book.

"The Essence of Chaos" by Lorenz, who was one of the pioneers of the subject, and he decided to write something that was understandable by outsider.

Just reading that book and having a grip on chaos theory will not make you a dime, but it might make you feel better in a hard to describe way.

I have embraced the concept that the markets are a chaotic system sort of like the weather. It offers an intellectual argument that the markets are not a random walk, but at the same time explains why there is no simple system that is the holy grail.
 
K guys, thx. It sounds like what is called 'chaos' in this context is really shorthand for 'hard to predict order.'

It still screams of order, albeit multidimensional, which is really my interest because, as you know, I'm always arguing that there is a Designer.
 
Chaos Theory is really just that - a pretty name. It's actually a study of a general class of differential equations.

Fractals, wavlets, etc.... are just that. Interesting little toys that are useful when it comes to natural science.

The market is not natural. The market is social. Popper's natural science method of understand (perdiction=explaination, explainations = perdictions) does not hold water here.

The market is stochastic. No, not the TA indicator, the math concept. It is by definition stochastic (nevermind what order and what degree).

Gordon Gekko is a character in "Wall Street." He failed. He lost. He was based loosely on Milkin and Boesky. They were traders, but they didn't trade stocks or bonds or whatever - they traded information. Information is not chaos. Information is not fractal.

Gordon Gekko is a character who failed.
 
I beg to differ about making money using non-linear dynamics.Trading is all about continued learning (about yourself and the markets in general).Many automated (Neural network,fuzzy logic, CA,or just plain vanilla charts) systems are used by many hedge funds, mutual funds ,pension funds, and banks to control inventory as well as new buys.As for chaos theory, yes you do have to have a talent for math and computer programing. But I seem to recall an idol of mine as saying " imagination is more important than knowledge". Something he also said was "god doesn't roll dice" about the non-deterministic hypothesis that is quantum physics. Can an experienced trader not tell when to buy the dips or sell the rallies? A good trader adapts to what he ( or she) see's with the measure of success being making money whereas your local weatherman forecasts based on the same principles. metrologist forecast based on computer models of chaos theory (a tornado has no predictability witch makes it chaotic but has structure within ,otherwise chaos would feed on itself and dissipate just after it is formed). R.N. Elliot saw the chaos in the markets short term but saw the order in the bigger picture as time expands.Can we not look at a chart and say it's overbought or oversold and ready to reverse,only needing a reason (econo number , earnings number or political garbage)and having it happen? Determinism is there (might want to look at steven wolframs new book) you just need the imagination to lead the way and test,test,test. A real passion for the market is all it takes (might also want to see the dvd "PI")and making money will come .
 
Originally posted by FasterPussycat
GG sadly it just ain't gonna happen for you 'ol buddy. you should be fervently studying charts etc and here you are wasting your time talking about fractals.
More...
FP,

i KNEW i should have also said this in my original post...but.....

my point for starting this thread wasn't really to find a way to make money with the concept. it was more of an effort to expand my general knowledge. i'm very interested in the universe/science.

p.s. i have stared at many, many, many charts....and yes, it is going to happen for me.
 
Originally posted by chasinfla
Would someone care to, again, please define 'chaos' for the purposes of this discussion? Thanks.
More...
chas,

go read my "markets are not random" thread. my original post explains it all.. heheeee
 
Originally posted by Gordon Gekko

chas,

go read my "markets are not random" thread. my original post explains it all.. heheeee
More...

And you deign to post to humble me? :D
 
Maybe the real cheesecake is going to come from CA. Look at what it is; simple rules that lead to complex behavior that show different outcomes with perturbation.If one cell's neighbor is a different color then it becomes that color. Now if there is a block of 5 cells and 2 are white and 3 are black and run the program for a few seconds,minutes,hours,or days you will see something extremely complex in structure from these simple rules. Watch it as it is running and you will say it is chaotic. But in the end a pattern emerges that has structure.Now do a filter and plot an analog of the running difference of the numbers of black and white cells at successive steps and see what it looks like; a chart pattern. Now apply this and make one cell the S&P 500 another say the 10 year not ,another the put/call ratio, etc. Apply the data in real time with white to be up and black is down ,as well as over 1,5,15,30 ,60 minute time frames and see what you get. Obviously 64-bit computing comes into play but would the rewards be worth it? Remember , the most successful systems are never told only discovered.
 
You must log in or register to reply here.
Back
Top