Thread: Random Price Movement

Victor Niederhoffer wrote an interest in article. Although he used the SP 500 since the instance it would evenly aply into FX IMO.

http://www.dailyspeculations.com/vic/momentum.html
http://www.dailspeculations.com/vic/momemtum.html

This theory is introduced in virtually every finance class as a testament to the efficient market theory. Yet as professors and professionals parade about asserting there is no point in calling the direction of this market but only the probabilitythey appear to overlook an integral factor; markets and coins, while they could paint similar pictures on a chart, are very different. Coins don't collect and disperse rankings therefore creating levels of support and resistance, this really is the gist of technical analysis the best traders in history have employed to make their fortunes. Perhaps people should stop being so concerned about what their super-ema. Rsi.macd.mega-oscillator states and begin paying more attention to the accumulation and supply of this market. Perhaps then it would stop appearing like a coin toss?

Originally Posted by ;

This theory is introduced in almost every finance class as a testament to the efficient market theory....

Correct, judging buyers / sellers will be the way to go.

The market does not look like a coin toss that would look like a pretty much directly wibbly lineup, together with generally ( depending on the coin ) that a slight edge towards the Heads side due to more mass landing so a slight trend caused by that.

Originally Posted by ;

Correct, estimating buyers / sellers is the thing to do.

The market does not appear to be a coin toss that would seem to be a pretty much directly wibbly line, together with generally ( depending on the coin ) a small advantage towards the Heads side because of more mass landing down so a small trend brought on by that.

Spoken like someone who didn't bother to conduct the experiment. There may be tendencies and patterns in random events as well as for the record, electronically simulated coins do not have mass imbalances.

Originally Posted by ;

So, what you're getting at here is the same patterns we see in charts that traders use to base trading decisions on, can be reproduced at random and there is not any underlying significance to such patterns?

That is 100%, without a doubt, certainly the point I am attempting to make.

I do not agree with that. It is to be anticipated a charted random input, like a coin flip, will have segments that match charting patterns. There is nothing surprising about this. In the fullness of time, that is an infinite number of trials, every last imaginable pattern will probably be current.

The important question is whether a specified pattern will look as often with an entirely random input as it will in market history. And more importantly, whether the supply of charts after a specific pattern appears in the markets, match exactly the same frequency as the supply of charts after that pattern comes up from charting a coin flip. As an example, a number of discretionary breakout traders look for pennants. They conclude that the market has come to some place where it has to fall out of equilibrium, and once it does there is a greater than anticipated probability it will last in 1 direction. Five straight up moves, with a 50/50 random inputsignal, will happen once in half two trials. Can a breakout of a pennant have exactly the exact same probability of a strong move in 1 direction, or does it have a greater probability? Curtis Arnold's response to the question, in Pattern Probability Strategy, is that pennants in the market have a greater probability. That 1 set of random inputs looked like a head and shoulders pattern doesn't tell us anything.

If a few charting patterns look more often than their anticipated frequency, then suggests that they do possess some inherent significance.

mark here!

Originally Posted by ;

Wow... what a great group of silliness we have here... sounds like a couple people who agree with one person wanting to ensure nobody else disagrees with them. Now that is what net forums are all about, right? Huh.

OK... I am a mathematician. Take my word for it. I suppose I should be one of the guys who believes deeply in these matters as mathematical randomness in markets. And I tip my hat to the core first point of this thread, that (if I follow), was'Isn't it interesting that randomness can regularly create patterns that...

ha. Great!

Originally Posted by ;

An interesting post. Thanks for taking the time.

I am making an assumption here so correct me if I am mistaken. You exchange a few market inefficiency that you've discovered? I can respect this. It is a wise approach to exchange.

You are probably familar with Edward O. Thorp. For those which aren't here is an interesting link about him:

http://webhome.idirect.com/~blakjack/edthorp.htm

Thorp transactions inefficiencies in the market. Really, really nicely in fact.

What I needed to point out about Thorp is his work BlackJack (another zero-sum game). He can...

Professional man!
If you take the mathematics finance courses. There is one course to take care of the random number generater!

Originally Posted by ;

The Excel Code appears impressive, but the arbitrary numbers a computer generates, are not truely random there psuedo random, which I think is skewing the data.