Analysis of outcomes in relation to intervention requests for unknown disease
This project looks at anonymised data for interventions and outcomes for an unknown disease, The hypothesis is: The longer it takes for the given intervention in an outbreak the more cases, people admitted to hospital and deaths there will be.
The biggest problem was the lack of data, which made it hard to get meaningful results, both in terms of having limited observations and having limited features to work with. The data are also discreet.
The analysis uses linear regression, poisson regression and negative binomial models.
Analysis.R contains all of the code.