We (ie. @tobylunt) made a district-level COVID forecast for India: devdatalab.org/covid-forecas…
A lot of places are past the peak of the second wave. In a lot of places, case counts are still rising.
We show Rt calculated from a model based on Bettencourt and Soman (2020). 🧵1/3
Are night lights (NL) a reliable proxy for local GDP?
An answer, in our paper “Development Research at High Geographic Resolution” aka the SHRUG paper (forthcoming at WBER). 🧵1/14
TL;DR: don’t trust existing GDP elasticities. With @paulnovosad @tobiaslunt @ryu_matsuura