Proof of a conjecture of Sturmfels, Timme and Zwiernik

TL;DR

This paper proves a conjecture regarding the ML-degrees of linear covariance models in algebraic statistics, utilizing intersection theory on varieties of complete quadrics.

Contribution

It provides a rigorous proof of a conjecture in algebraic statistics, advancing understanding of ML-degrees in linear covariance models.

Findings

Confirmed the conjecture on ML-degrees of linear covariance models.

Connected the proof to intersection numbers on varieties of complete quadrics.

Extended previous work on linear concentration models.

Abstract

We prove a conjecture of Sturmfels, Timme and Zwiernik on the ML-degrees of linear covariance models in algebraic statistics. As in our previous works on linear concentration models, the proof ultimately relies on the computation of certain intersection numbers on the varieties of complete quadrics.