Project
Bayesian Psychometric Network Modelling for Longitudinal Data

My Ph.D. project is part of the ERC-funded Bayesian P-Nets project, headed by Maarten Marsman, which aims to develop new Bayesian methods for analyzing psychometric network models (i.e., undirected graphical models, also known as Markov random fields (MRFs); Kindermann, & Snell, 1980). This subproject aims to develop new autoregressive MRFs for longitudinal data and Bayesian edge selection and structure averaging methodology to analyze them (e.g., Madigan, & Raftery, 1994; Hoeting, Madigan, Raftery, & Volinsky, 1999). This approach accounts for the uncertainty of estimated parameters of any temporal structure and the uncertainty of selecting the structure itself, yielding a novel framework for exploratory and confirmatory longitudinal network modeling. 

The basis for our autoregressive MRF models is the cross-sectional MRF models that are studied and developed in another subproject of the Bayesian P-Nets project (see IOPS application of Nikola Sekulovski). These include MRF models for the binary and ordinal data common in psychology and extend the popular Gaussian MRFs for longitudinal data (e.g., Epskamp, 2020). 

The Ph.D. project comprises four subprojects:

1. Develop prior specifications for a longitudinal network’s parameters and architecture in psychological contexts.
2. Explore and develop network models for longitudinal data that cover different types of variables (e.g., binary, ordinal, and continuous) and estimation methods for their analyses.
3. Explore and develop structure averaging techniques that overcome the difficulties of model-averaged assessment of network models for longitudinal data (e.g., Occam’s window and stochastic search methods).
4. Develop Bayesian methods for evaluating hypotheses about longitudinal data (e.g., critical transitions).

In my view, the Ph.D. project outlined above fits the IOPS graduate school since it explores and develops methods for psychometric network models and psychological data.

Supervisors
Prof. dr. H.L.J. (Han) van der Maas  
Dr. M. (Maarten) Marsman

Financed by
The European Research Council

Period
1 September 2022 – 1 September 2026