-2
votes
0answers
21 views

Can a very bad Coefficient of determination (R squared value) not be indicative of model performance? [migrated]

Thanks in advance for the advice. I am trying to build a generalized linear model that has many predictors. The R squared value of the model is quite low (.21), but when I use the model to predict ...
9
votes
1answer
310 views

Coherence between different ranking methods of a graph's vertices

Given a (connected) graph $G$ it is natural to want to rank its vertices, with the more "central" vertices ranked higher. Two natural ways of doing it are: By the degrees. By the entries in a ...
2
votes
0answers
110 views

Optimization over Spectral Laplacian in cycles and trees

Is there any idea on how one can deal with an optimization problem of sum of k largest eigenvalues(min) of Laplacian matrix of a simple cycle or tree? I would like to use semidefinite programming for ...
0
votes
0answers
265 views

Orthogonal Projections in Lie Theory

I have been studying a finite element method where rigid & elastic spatial motions are separated using an orthogonal projection (actually two: one for translations/stretches, the other for ...