commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Phil Steitz <stei...@yahoo.com>
Subject [math] Reference data
Date Fri, 23 May 2003 00:18:44 GMT
I am using the linear regression reference datasets here:
http://www.itl.nist.gov/div898/strd/general/dataarchive.html
to test the "no memory required" (similar to
UnivariateImpl) Bivariate regression implementation that I
am preparing for submission.

There are also datasets there that can be used to test the
Univariate implementations.  I like the idea of using
reference data sets and verified statistics in our unit
tests.  The archive above is limited, but I am sure that
there are lots of others available.  The nice thing about
the nist archive is that the precision of the statistics is
provided.

In the stuff that I have submitted up to now, I have been
relying on either other software packages or examples from
books to provide validated data for the tests.  Moving
forward, I suggest that whenever possible, we try to use
publicly available reference data and validated
computations in our unit tests.

Phil

__________________________________
Do you Yahoo!?
The New Yahoo! Search - Faster. Easier. Bingo.
http://search.yahoo.com

---------------------------------------------------------------------
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