Using Chi-Sq when we don't use least squares

ATTPC / Spyral

A Python analysis library for AT-TPC data

GNU General Public License v3.0

2 stars 2 forks source link

Using Chi-Sq when we don't use least squares #87

Closed gwm17 closed 5 months ago

gwm17 commented 5 months ago

We save a reduced Chi-sq which is confusing, as we don't use a least-squares approach anymore

gwm17 commented 5 months ago

On second thought this isn't such a bad thing. The reported value when using L-BFGS-B is the residual (objective function value) squared. If people want to modify that they can (i.e. take the square root) but I don't think it does much to change the story and it allows us to compare models to things that do report a true least-squares. Marking as closed for now.