Description The Covariance calculation is useful in determining if two series of numbers are related.
SUM ((Xi - mean of X) - (Yi - mean of Y)) / n
For two symbols X and Y, if their time series are not closely related then the covariance will be small,
and if they are simliar then the covariance will be large.
Two processes are defined to be "closely related" if their distribution spreads are almost equal and they are around the same,
or a very slightly different, mean.
Parameters:
BarsBack Period with which to base the Covariance calculation
Arguments:
OHLC of the Time Frame, or an
Output Indicator of a study in the TimeFrame.
Symbol - A valid symbol is the control series which is compared against
all the other symbols currently being scanned
Output Indicators:
CoVariance
Example:
Study Name Expanded in a 3 minute timeframe:
I3_CoVariance(10)(#SPX#)_I3
This study calculates a Covariance with a period of 10 on the 3 minute time frame, using
#SPX# as the control symbol (#SPX# is the Stormtracker symbol for S&P 500 Index).
The output indicators names are appended to the studyname,
that is if the studyname is sn1 then
the outputindicator is
sn1::CoVariance
Home | What's new | Purchase | Download trial | News | Tech Support | Contact Us | Disclaimer
Copyright © 2000. All Rights Reserved. Storm Chaser Technologies
