Back

A Transparent, Open-Source Sird Model For Covid19Death Projections In India

Agarwal, A.; Tyagi, U.

2020-06-04 epidemiology
10.1101/2020.06.02.20119917 medRxiv
Show abstract

AO_SCPLOWBSTRACTC_SCPLOWAs India emerges from the lockdown with ever higher COVID19 case counts and a mounting death toll, reliable projections of case numbers and deaths counts are critical in informing policy decisions. We examine various existing models and their shortcomings. Given the amount of uncertainty surrounding the disease we choose a simple SIRD model with minimal assumptions enabling us to make robust predictions. We employ publicly available mobility data from Google to estimate social distancing covariates which influence how fast the disease spreads. We further present a novel method for estimating the uncertainty in our predictions based on first principles. To demonstrate, we fit our model to three regions (Spain, Italy, NYC) where the peak has passed and obtain predictions for the Indian states of Delhi and Maharashtra where the peak is desperately awaited.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.