COVID Prevalence Estimate
An implementation of Bayesian inference and prediction of COVID-19 point-prevalence.

Definition of point-prevalence:

The portion of infected individuals in a population at a given time.

Model Description

The model is an SEIR-like model, which treats reported cases as a biased partial measure of symptomatic onset with a variable delay to test and report. Consideration is given for the likelihood of asymptomatic cases and undetected cases, as indicated from the current epidemiological parameter estimates. The model debiases the case data for reporting and epidemiological effects using the entire data hysteresis to provide an estimate of the true prevalence within a population. Since the focus of this model is on active (infectious) cases in the population, recovery in quarantine and deaths are not considered.

Once the model is tuned to the case detections of the population, a prediction is provided to provide an indication of the expected future prevalence and cases.

Acknowledgements

The following individuals have provided support or contribution to this project:

Dr. Ramzi Mirshak
Dr. David Waller
Dr. Steven Schofield
Mr. Michael A. Salciccoli
Dr. Steve Guillouzic
Mr. Andrew Sirjoosingh
Mr. Alasdair Grant

Immediately related elementsHow this works
-
EIDM  »EIDM 
Software »Software
Software programs »Software programs
COVID Prevalence Estimate
Python »Python
Fitting dynamical models to data »Fitting dynamical models to data
Compartmental »Compartmental
Stochastic »Stochastic
Steve Guillouzic »Steve Guillouzic
Matthew MacLeod »Matthew MacLeod
Covid-19 »Covid-19
Bayesian analysis »Bayesian analysis
+Komentarai (0)
+Citavimą (1)
+About