Statistical reliability is not to be confused with the probability of an outcome. A statistical reliability is saying this or that will happen, a probability of a statistical reliability is a certain.....so a probability must have any number of outcomes...Correct? Trading is a probability business. i believe the best statistical indicator going is your intuition. it is refined every time you recognize a mistake and further refined every time said mistake is corrected. thus you become automatic in pulling the trigger. Fear is eliminated from the game. Gotta go buy a new TV.....for another room.....
nice article about Niederhoffer. what he didn't understand is that markets obey powers laws and not gaussians. power laws are fascinating, they are the distributions for scale free entities like markets. in other words they look the same in most time scales (like price charts can) and have 'fat tails' http://polymer.bu.edu/hes/articles/ymhbs05.pdf empirically speaking, it has been found that the stock market is correlated for about 20 minutes with respect to price direction.
sure, enjoy but please give opinions. Econophysics seems to be developing as a discipline but I have to wonder how much the PhDs at Goldman know that they can't talk about. This stuff is of more use to options traders than to straight position traders, but I do think it helps to understand the historical distribution of fat tail events. I find it fascinating that although things like 9/11 are man made events, we can still talk about the probability distribution function of such things in the stock market. This is not to be taken as predictive though, we are just looking at averages.
Wow, great thread. I'm quoting this comment out of all the great ones, because doesn't this hit the nail on the head? You mention things (the market and others) are random, but here you state that this thread will surely be referenced? Isn't there at any post the possibility that it will be the last and the thread will die and fade into obscurity? I would offer up that the markets are not random like coin flipping, but rather behave more like this message board. Within reason there is no predictable outside force that can swing the outcome of a coin toss. However, the markets and the posting and reading of a message board are dictated by human behavior. Human behavior is predictable based on previous observations. One could build a system around knowing that if X number of posters are engaged in a discussion, there are open questions, and open disagreements there will be a very high probability that the thread will continue for at least another post. The markets, in my opinion work the same way. I know that when a piece of news comes out there is a very high probability that the particular stock in question is going to move in a certain direction. There will be the initial move of those that follow closely and there will be a trailing move as those that are late to the party decide that it's time to get out or it's time to buy more. I also know that a particular stock moves out beyond it's fundementals in relation to similar stocks there will eventually be some sort of reconciliation. There are too many people that trade on fundementals to let things remain out of wack too long. The same could be said for certain sectors. We know that people play trends (fact) and because there it is human behavior that interprets the trend we can see how people have behaved on previous trends in the past. I could give more examples, but in summary, the market and thus the charts are a result of human beings making decisions and human behavior is predictable. There are unpredictable events, but how the market behaves in response is predictable. Mury
Traders must understand the law of probability and Heisenbergs uncertainty principle... in short... I cant predict the outcome of a single coin toss... but I can predict what it will do over 100 tosses... dont get caught up on the single trade, it's a mystery and will always remain so... know what is predictable and what is not
"Treating the market as a mechanism for testing hypotheses seems to be an effective hypothesis. It produces results that are better than a random walk." - G Soros.
would offer up that the markets are not random like coin flipping, but rather behave more like this message board. Within reason there is no predictable outside force that can swing the outcome of a coin toss. However, the markets and the posting and reading of a message board are dictated by human behavior. Human behavior is predictable based on previous observations. Human behavior is not predictable, i disagree. the events that drive human behavior are not predictable, nor is the exact response to those events. hurricane Katrina and 9/11 are examples. your own response to a given event may differ depending on all circumstances. sometimes responses though are unform to a high degree I'll have to admit. If we have a surprise freeze in July in Iowa- Corn's going to be bid up 99% of the time.