search

epiverse-trace / vaccineff

R package with tools for estimating vaccine effectiveness and vaccine related metrics
https://epiverse-trace.github.io/vaccineff/
Other
14 stars 2 forks source link

Dynamic matching #37

Closed fsabrilb closed 3 months ago

fsabrilb commented 9 months ago

In order to apply the semiparametric Cox model of vaccine effectiveness, it is necessary to make a population match of the study population according to some covariates such as age and sex. For this reason, it would be useful to apply a population matching method on the cohorts that is dynamic over time in order to have scenarios that are closer to what is required for the Cox model.

davidsantiagoquevedo commented 9 months ago

@fsabrilb I replaced the example dataset with a new one that includes a date column for death by other causes (death_other_causes) to work with censoring, and an anonymous ID (id) to match the cases. I will have to compress this data in the future, but for now, it is enough to work on the examples for dynamic matching. Please install the package from the branch associated with this issue using: pak::pak(epiverse-trace/vaccineff/tree/feat-censored-data)

davidsantiagoquevedo commented 3 months ago

Solved in https://github.com/epiverse-trace/vaccineff/pull/57