Statistical Inference on Interacting Particle Systems
Licentiatavhandling, 2021

Interacting particle systems, and more specifically stochastic dynamical systems, is a mathematical framework which allows for condensed and elegant modelling of complex phenomena undergoing both deterministic and random dynamics. This thesis is concerned with the topic of statistical inference on large systems of interacting particles, with the specific application of in vitro migration of cancer cells. In the first of two papers appended with this thesis, we introduce a novel method of inference based on a higher order numerical approximation of the underlying stochastic differential equations. In the second paper, we formulate a model for glioblastoma cell migration, and conduct inference on this model using microscopy data. This regression shows promising results in its predictive power.

stochastic process

mathematical biology

agent based modelling

glioblastoma

Bayesian inference

MVL15
Opponent: Linus Schumacher

Författare

Gustav Lindwall

Chalmers, Matematiska vetenskaper, Tillämpad matematik och statistik

A conjugacy for isotropically diffusive particle systems

Inference on an interacting diffusion system with application to in vitro glioblastoma migration

Infrastruktur

C3SE (Chalmers Centre for Computational Science and Engineering)

Styrkeområden

Hälsa och teknik

Ämneskategorier

Bioinformatik (beräkningsbiologi)

Sannolikhetsteori och statistik

Utgivare

Chalmers tekniska högskola

MVL15

Online

Opponent: Linus Schumacher

Mer information

Senast uppdaterat

2021-05-28