UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Non-parametric Bayes in biostatistics

Tallarita, Marta; (2021) Non-parametric Bayes in biostatistics. Doctoral thesis (Ph.D), UCL (University College London). Green open access

[thumbnail of TallaritaPHD.pdf]
TallaritaPHD.pdf - Accepted Version

Download (6MB) | Preview


The main focus of this Phd project is the application of Bayesian models in Biostatistics.It has become indeed evident that healthcare management is in need for methods able to improve evidence-based practice. The first problem we consider is modelling recurrent event and survival time. Recurrent event processes generate events repeatedly over time and they arise in many applications.Typically, the focus is on modelling the rate of occurrence, accounting for the variation within and between individuals. Moreover, in applications, it is often of interest to assess the relationship between event occurrence and potential explanatory factors. Although the first focus of our work is on modelling the recurrent event process itself, we also extend the proposed model as building block in a hierarchy to describe the relationship between recurrent events and survival up to a terminating event. This is achieved by specifying a joint distribution of the gap times and event (termination) time. The second objective is to identify the most promising methods that can be applied in a network meta-analysis (NMA) across longitudinal time points, compare them and extend existing models in a B-spline setting. The network meta-analysis methods extend the standard meta-analysis methods, allowing pairwise comparison of all treatments in a network in the absence of head-to-head comparisons. We focus on the most recent methods suggested in the literature that incorporate multiple time points and allow indirect comparisons of treatment effects across different longitudinal studies. In particular, we compare the Mixed Treatment Comparison model (MTC) Dakin et al. (2011), the Bayesian evidence synthesis techniques — integrated two- component prediction (BEST-ITP) developed by Ding and Fu (2013) and the more recent method based on fractional polynomials of Jansen et al. (2015). After a comparison of these methods, we develop some models within a B-spline framework.

Type: Thesis (Doctoral)
Qualification: Ph.D
Title: Non-parametric Bayes in biostatistics
Event: UCL (University College London)
Open access status: An open access version is available from UCL Discovery
Language: English
Additional information: Copyright © The Author 2021. Original content in this thesis is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Licence (https://creativecommons.org/licenses/by-nc/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request.
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences > Dept of Statistical Science
URI: https://discovery.ucl.ac.uk/id/eprint/10131027
Downloads since deposit
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item