On 9/7/10 9:24 PM, pste...@apache.org wrote:
Author: psteitz
Date: Wed Sep 8 01:24:51 2010
New Revision: 993574
URL: http://svn.apache.org/viewvc?rev=993574&view=rev
Log:
* Modified multiple regression newSample methods to ensure that by default in
all cases,
regression models are estimated with intercept terms. Prior to the fix for
this issue,
newXSampleData(double[][]), newSampleData(double[], double[][]) and
newSampleData(double[], double[][], double[][]) all required columns of "1's"
to be inserted into the x[][] arrays to create a model with an intercept
term;
while newSampleData(double[], int, int) created a model including an
intercept
term without requiring the unitary column. All methods have been changed to
eliminate the need for users to add unitary columns to specify regression
models.
* Improved javadoc
* Improved tests
JIRA: MATH-411
I agonized over handling the behavior change here via deprecation +
adding new methods. If the old behavior was documented, I would
have done that; but when you combine the fact that the documentation
was so poor (still needs work, but it is better now) and the
non-trivial tests all use the "correct" (intercept included) method,
I thought it was best to make the behavior consistent, document the
change in the release notes, and show how to get the no intercept
behavior once MATH-409 is fixed. If anyone feels strongly that we
should go the add new methods and deprecate route (making an
arguably already overly complicated API even more complicated), I
can revert this change and do that for the 2_x branch.
Phil
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org
