Description: Requires skills in R
and may require, in some instances, some learning on Bayesian modelling. The project can be split into a couple of different sub-projects that different students can take on at the same moment. Students are encouraged to work with Rmarkdown
or quarto
to develop their dissertation.
-
Low/medium difficulty π¬ or π¬ π¬: Meta-analysis of vaccination against COVID-19
- This involves the analysis of data collected through a systematic literature review, pooling together studies involving relatively homogeneous clinical outcomes, for a single disease outcome (e.g. respiratory disease).
-
Medium/high difficulty π¬ π¬ or π¬ π¬ π¬: Bayesian Meta-analysis of vaccination against COVID-19
- This expands on the above project to perform a Bayesian version of the modelling
-
High difficulty π¬ π¬ π¬: Bayesian Meta-analysis of vaccination against COVID-19 of heterogeneous clinical outcomes/disease areas
- This expands on the above projects to perform a Bayesian version of the modelling and considering outcomes that are not measured on the same scale (e.g. odds ratios, hazard ratios and relative risks), as well as across different disease outcomes (e.g. mortality and respiratory disease).
-
High difficulty π¬ π¬ π¬: Bayesian spatio-temporal modelling to estimate the excess mortality due to COVID-19
- This is a more complex modelling exercise, using population data (eg Census) to model the number of deaths in a time-series; predict the expected mortality during the pandemic, using past records; and then comparing the observed mortality with the predicted values, to estimate the excess due to Covid-19.