[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Article in Omega Research magazine.



PureBytes Links

Trading Reference Links

At 11:52 AM 9/8/99 -0700, Dave DeLuca wrote:
>In the last issue of OR mag (Fall 1999) there is an article titled
>"Calculating Stops with Statistics". On page 43 the author,
>referring to a daily bar, makes the statement
>
>"In accordance with the rules of statistics, and assuming a normal
>distribution, approximately two-thirds of all outcomes will lie
>within one standard deviation (plus or minus) from the mean."
>
>She is apparently referring to the trades that occur within a bar.
>
>My question is, is there any reason why we should assume a normal
>distribution? This doesn't seem logical.
>
>Dave

It makes the analysis easy. It fits a lot of real-life situations.

However, in finance, the distributions are frequently found to be 
"fat-tailed." So the assumption for trading is questionable.

It's similar to the way linearity assumptions are frequently used for 
analysis. The math is easy but the applicability to real-life may not be there.