Monte Carlo study of approximate confidence regions in some Gaussian branching processes
Title: Monte Carlo study of approximate confidence regions in some Gaussian branching processes
SNIC Project: LiU-compute-2021-1
Project Type: LiU Compute
Principal Investigator: Hao Chi Kiang <hao.chi.kiang@liu.se>
Affiliation: Linköpings universitet
Duration: 2021-01-08 – 2021-07-01
Classification: 10615
Keywords:

Abstract

In comparative phylogenetic analysis, multivariate Gaussian branching processes is a class of useful models to infer about the evolution of traits of species. However, currently it is not clear how to obtain a confidence region efficiently. I have found some methods to approximate the confidence regions by estimating the Fisher's Information, and this project aims at evaluate, via Monte Carlo methods, the how good my approximations are.