BAMresearch / probeye

A general framework for setting up parameter estimation problems.
MIT License
5 stars 5 forks source link

Improve general loglikelihood evaluation #71

Closed JanKoune closed 2 years ago

JanKoune commented 2 years ago

This pull request implements an improved version of the general loglikelihood evaluation, i.e. the loglikelihood evaluation when there is no structure in the covariance matrix that we can take advantage of. The direct inversion of the covariance matrix is replaced by a Cholesky factorization approach, which offers the following advantages: