Events
NDMC Talk Series: Invited Talk by Dr. Niket Thakkar, Scientist, Institute for Disease Modelling
November 13, 2025
Dr. Niket Thakkar visited NDMC and also delivered a talk on "Time and space correlations in measles surveillance systems"
Dr. Thakkar is a Principal Research Scientist at the Gates Foundation's Institute for Disease Modeling. A mathematician by training, Dr. Thakkar's current research focuses on developing and applying quantitative methods to open questions in immunization and vaccine delivery. In his role at the Gates Foundation, Dr. Thakkar has worked closely with a variety of global partners including the World Health Organization, the Vaccine Alliance, the US Centers for Disease Control and Prevention, and a number of governments. In those collaborations, his research has been used to design and advocate for immunization efforts that increase vaccine access and limit the burden of preventable diseases. Generally speaking, Niket's mathematical interests range across problems in probability, discrete math, and stochastic processes, and his research career has touched on topics in nanoscale optics, statistical physics, information theory, signal processing, and epidemiology.
Abstract
In this talk I describe some recent work on the measles forecasting models currently used to inform vaccination programs and campaign tactics in multiple countries. A key component of these models is their assessment of measles reporting rates, the fraction of measles infections reported as cases to the health system, and by extension the amount of transmission happening cryptically in the population. I'll discuss the regularization approaches underpinning reporting rate estimation in these forecasting models, exposing reporting rate variation as an important component of the epidemiology, one with unusual and intricate noise properties. By way of example, I'll demonstrate that forecast accuracy depends strongly on our understanding of these noise properties, and with that motivation, I'll conclude by describing some current work on characterizing time and space correlation functions of measles reporting rates.