[SciPy-User] linear algebra: quadratic forms without linalg.inv

Gael Varoquaux gael.varoquaux@normalesup....
Sun Nov 1 21:06:52 CST 2009


On Mon, Nov 02, 2009 at 04:03:14AM +0100, Sturla Molden wrote:
> josef.pktd@gmail.com skrev:
> > In econometrics (including statsmodels) we have a lot of quadratic
> > forms that are usually calculate with a matrix inverse. 
> That is a sign of numerical incompetence.

Yup, but you'd be surprised to see how much inverse is used. I was
astonished to find out that senior researchers that I respect a lot were
not even aware of the problem.

Gaël


More information about the SciPy-User mailing list