|
PureBytes Links
Trading Reference Links
|
----- Original Message -----
From: "Brent" <brente@xxxxxxxxxxxx>
To: "Omega List" <omega-list@xxxxxxxxxx>
Sent: Friday, January 10, 2003 4:15 PM
Subject: RE: FW: Random walk
: > If you doubt it, just tell me what the exact price of the Dow will be 5
: days from today at 9:00est...
Brent, all this shows is your lack of knowledge. The exact Price of the Dow
Jone Industrials 5 days form today at 9 am EST is the same price it closes
at the night before. The US stock markets don't open until 9:30 am EST.
: Excellent job, but if in the end result the system is so chaotic, I'm not
: sure that it could be proven. We really need a direct and simply elegant
: proof.
Take the log(price) - log(price)[1] of the S&P 500 and find the distribution
that fits it. I can assure you it's not the Gaussian/normal distribution but
a Stable Pareto Levy distribution (which means that the markets are
'predictable' in the short term but never in the long term.). No way in the
world it's random.
Depending on what formula you use, the kurtosis of the S&P 500 is never zero
(it may becomes zero momentrarily as it crosses above/below it). If the
kurtosis is zero, you'd have a random market. On a related note, the same
applies to the skewness of S&P 500. No way in the world it's random.
Find the Hurst coefficient of the S&P 500. It may go above 1/2 or go 1/2 but
only becomes 1/2 in its meandering (1/2 means it's random, above 1/2 means
it's trending and below means it's in a trading range.).
Attachment:
Description: "snap.gif"
|