Stocks are random variables

Quote from WallStGolfer31:

Don't ever send me or anyone else a wikipedia link and quote from it if you want anyone to even think of respecting you.

And don't believe me over him, believe the 100's of thousands of Dr.'s past and present who have provided evidence for EMT.
More...

The most successful hedge fund manager of all time makes his money trading predictive patterns in the markets (from the horse's mouth) in direct contradiction to your assertions, and all you can do is complain about wikipedia?

You lose. Big. Over and out.

Fletch
 
Quote from fletch2:

The most successful hedge fund manager of all time makes his money trading predictive patterns in the markets (from the horse's mouth) in direct contradiction to your assertions, and all you can do is complain about wikipedia?

You lose. Big. Over and out.

Fletch
More...

Oh, I guess it's over since you have proclaimed yourself the winner almighty genius eh?

LMFAO, keep believing in your chart patterns and keep researching old company financials to determine stock prices. LOL just keep it up

(Trying to convince someone by showing them the wikipedia article about it? That's just plain retarded)
 
Quote from WallStGolfer31:

Oh, I guess it's over since you have proclaimed yourself the winner almighty genius eh?

LMFAO, keep believing in your chart patterns and keep researching old company financials to determine stock prices. LOL just keep it up

(Trying to convince someone by showing them the wikipedia article about it? That's just plain retarded)
More...

Do your own research, my time is not worthless. Here, I will waste a few more seconds of my life on you:

http://infoproc.blogspot.com/2005/11/simons-thorp-and-shannon.html

As Mr. Simons put it - and this is about as specific as he would get - "Certain price patterns are nonrandom and will lead to a predictive effect."

This is the same Dr. Simons that has posted 35% a year since 1998 after fees.

EMT is approximate and price is not fully random. That's the truth. Do with it as you will.

Fletch
 
Quote from whitster:


there is no more democratic institution on earth than an electronic order entry book.
More...
EXACTLY !
if you "buy" stock it means someone sold it to you, right?
if he sold it to you, it means that he don't beleive it will rise, right?
(if someone believe price will be higher, WHY he will sell stock to you???)
so if most people believe price will be higher then only way to buy stock is to bid for higher price, right? and "bid" at such level than someone stops to believe price will be even higher (if someone believe price will be even higher, WHY he will sell stock to you???)
so it's absolutely obvious that price is just a game between those who believe price will be higher and those who believe price will be lower.
If you seen LEVEL II quotes at least 5 minutes in your life you must understand that there is no "true" value for price, its just a war between "bid" and "ask"
and you can make money if you play that game better than "average joe", just like in sport and in war.
 
Quote from ashcroftsinger:

Based on this observation, you have two potential inferences(amongst many) - The coin may not be fair, and hence there is more than 50% chance that it will turn heads again, or the coin is fair, and the next throw will have equal chance of a head or a tail. IMO there is some value in the first opinion. After all, if theory is our best attempt at explaining observed phenomenon, why disregard such strong evidence just based on theory?

There are many sources of market inefficiencies, let me site one obvious one: Let us say, you are right. However, it is obvious that there are many people in the market who do not agree. That would lead them to make suboptimal decisions, which implies that you just have to take the other side to their trades to make money.

Now, can we identify these inefficiencies and turn them into a successful, consistent trading system? May be, may be not. Inability to find such trading methods does not prove that there are no inefficiencies.
More...

called the bi nomial distriution

With 15 in a row called Runs test..or another called Sign test
also used to test if some data series is random...from one point to the next
 
Quote from WallStGolfer31:

You keep telling me about these statistics but all you keep quoting is individual case studies?
More...
if we are talking about random walks his individual case study is absolutely significant, given that there is a high number of trials.
 
Quote from whitster:


the market is not PERFECTLY efficient
More...
actually, I'm not sure an efficientist would argue with that. I think someone would raise the point: is the market inefficient enough to let you make money? Or you will lose it anyway, because of slippage, commissions, etc?
Don't really agree with that, I just thought it was fair to point it out.
 
Quote from WallStGolfer31:

You can't predict the future with patterns on a chart, that's just absurd. You can't predict future prices by doing research based financial statements from the past, it's already priced in!
More...

I find some securities have a history of showing price change continuation, sometimes called trends.

Attached is a trading simulation of Walgreens stock (symbol WAG) using random entry and exiting position when price decreases to less than the lowest price of the prior 30 daily sessions. Risk to protective stop is 1 %. Simulation begins 1 July 1985 ends 24 Jan 2007.

Results are similar when I change the number of days for exit. General Motors stock shows similar results.

I view back testing as identifying a property of price behavior that might be called tendency to trend. Once a trend characteristic is identified a trend following method might be applied at any time and might show an overall profit.

I run 100 loops and sort the data from greatest profit to greatest loss. These are the results of 100 individual simulations.

I observe that many of the low profit test runs (the simulations near the center of the x axis on the graph) also show a low (less than 100) number of trades. I suspect profitability of these trades is low because the system does not always trade often. Computer random number generators might not always trigger trading frequently enough to show large profit.

Average profit is about $ 233,000, initial capital $ 100,000. Median profit is $ -200. The system shows greater than $ 100,000 profit in about 25 % of the simulation runs.
 

Attachments

You must log in or register to reply here.
Back
Top