Scalable Bayesian Inference for Comparative Judgement Models

Add to your list(s) Download to your calendar using vCal

• Rowland Seymour (University of Birmingham)
• Tuesday 08 November 2022, 14:00-15:00
• Watson R17/18.

If you have a question about this talk, please contact Hong Duong.

Forcing someone into a marriage against their will is a human rights violation. The county of Nottinghamshire, UK, launched a strategy to tackle forced marriage and violence against women and girls in 2021, but there is no high resolution spatial information of where victims of forced marriage are located in the county. Therefore, it is not possible to develop targeted interventions to support victims. Comparative judgement studies are being increasingly used for social good causes, and practitioners supporting victims and survivors of forced marriage are aware of locations in Nottinghamshire where forced marriage is a common practice. However there is a limited number of practitioners who can provide information and their time to take part in studies is often limited. In this talk, I will present a new set of comparative judgement models that provides a more flexible spatial modelling structure with an efficient mechanism to schedule comparisons. The methods reduce the data collection burden on individual participants and makes a comparative judgement study feasible with a small number of participants. Underpinning these methods is a P\’olya-Gamma latent variable method that improves on the scalability and efficiency of previous comparative judgement models. I am able to map the risk of forced marriage across Nottinghamshire and support the county’s strategy for tackling violence against women and girls.

This talk is part of the Data Science and Computational Statistics Seminar series.

This talk is included in these lists:

• Analysis Seminar
• Data Science and Computational Statistics Seminar
• School of Mathematics Events
• Watson R17/18

Note that ex-directory lists are not shown.