URV's Author/s: Wehrens, Ron Giordan, Marco

Keywords: Levenberg-Marquardt algorithm, re-parametrization, starting values, metabolomics data

Abstract: Likelihood estimates of the Dirichlet distribution parameters can be obtained only through numerical algorithms. Such algorithms can provide estimates outside the correct range for the parameters and/or can require a large amount of iterations to reach convergence. These problems can be aggravated if good starting values are not provided. In this paper we discuss several approaches that can partially avoid these problems providing a good trade-off between efficiency and stability. The performances of these approaches are compared on high-dimensional real and simulated data.

Journal publication year: 2015

Publication Type: info:eu-repo/semantics/publishedVersion info:eu-repo/semantics/article