commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Phil Steitz" <phil.ste...@gmail.com>
Subject [math] CorrellatedRandomVector - doco and component distributions
Date Sun, 20 May 2007 20:12:30 GMT
Are there any references that we can cite for the algorithm
implemented in this class?  Also, am I correct in assuming that the
covariance matrix effectively determines the variances of the
component distributions and their (individual, unconditioned)
distributions are Gaussian with the supplied means?  It might be good
to specify the formula for the component variances, which IIUC should
be a sum over some entries in the root matrix.  Could well be I am
misunderstanding the code and this would be messy, but if it is not,
it would be good to specify.  Just spelling out how the values are
generated using the root matrix and normalized Gaussian generator
would probably suffice.

It would also be good to add a small bit on random vectors to
random.xml in the user guide.

Phil

---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org


Mime
View raw message