|
PureBytes Links
Trading Reference Links
|
At 10:51 PM -0400 6/3/00, Ronald McEwan wrote:
>In "Option Volatility and Pricing Strategies" by Sheldon Natenberg, he
>makes a couple of statements about distribution of price data
>
>1.The percentage changes in a security's price are distributed normally
>
>2. The dollar changes in a security's price are distributed log normally.
These are the same thing, of course.
It is fairly well known that the distributions tend to be slightly narrower than "normal" in the middle of the distribution and that the "tails" of the distribution are "fatter" than "normal". In fact, several people have shown that the variance (square of the standard deviation) can be infinite.
The major implication is that highly unlikely events as determined by the "normal" distribution will occur far more frequently than predicted from the "normal" distribution.
All of that being said, for most things the assumption of a "normal" distribution will give you the correct results for most purposes, especially within two standard deviations of the mean.
Bob Fulks
|