[RT] Stock Fingerprints

[Ronald McEwan <rmac@xxxxxxxx>]

I have been working on ways to identify the individual characteristics of
tradable securities (stocks and indexes). With all that has been said
about distribution of prices and price changes, I think much of this
amounts to over generalizations. I think each security has it's own
unique identity, like a set of fingerprints. Only with securities the
fingerprints can change over time. So it is important to monitor the
changes if we want our systems to continue to perform well. First I
compared the curves of a normal Gaussian distribution (based on the mean
and standard deviation of its daily returns) with the actual distribution
of the daily returns. In this example I used CSCO (for the last 100
days). CSCO has a unique pattern of its actual 100 daily returns versus
the expected normal distribution curve. It spent more time then expected
trading up and down approximately 5%. And it trades up and down 10% more
times then the normal curve would predict. I use the normal curve as a
comparison because the Black Scholes model is based on a normal
distribution curve. This analysis helps to explain the "Volatility Smile"
where out of the money options can sometimes be priced higher then the
Black Scholes model would predict. More importantly I wanted to see what
happens to CSCO when these "more then expected" events happen. As you can
see from the chart these events signal significant changes in the
direction and magnitude of CSCO's price changes. As I mentioned the
purpose of this was to identify the particular "fingerprints" of a
security in able to apply a suitable trading strategy and as a filter for
picking securities to trade.

Ron McEwan. 
To unsubscribe from this group, send an email to:
realtraders-unsubscribe@xxxxxxxxxxxxxxx

 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Attachment: Description: ""