Re:An Introduction to Gradient Descent and Linear Regression
by grayclhn on Thursday June 26, 2014 @03:22PM
(#7951173)
Just to add to the other comments: you almost never want to invert the X'X matrix. (Which is k by k, where k is the number of regressors.) The QR decomposition is much more reliable. (for those that want a citation: Seber and Lee's Linear regression analysis is great.)
[ Parent
| on HN ]
There are no comments on this comment yet.