Natesh Pillai is an associate professor of statistics at Harvard University. He is an interdisciplinary scientist working on problems lying at the interface of applied probability and statistics, with a particular research focus on Markov Chain Monte Carlo algorithms and Statistical Inference for high dimensional problems.
From his website: "My broad interests and genuine curiosity in different fields of mathematics and statistics gives me a chance to talk to and learn from other scientists. Applied mathematics is concerned with developing models which yield both qualitative and quantitative insight into the physical phenomena being modeled. Statistics is data-driven and is aimed at the development of methodologies to understand and account for the uncertainty in the information derived from the data. The main motivation for my research is the prospect of combining the increasingly complex physical phenomena that scientists and engineers observe and wish to model, together with the plethora of modern statistical techniques developed to understand uncertainty. This exciting aspect requires these two subjects to work in conjunction in order to significantly improve our knowledge with beneficial impacts on both the disciplines."
Talk: Bias correction in daily maximum and minimum temperature measurements through Gaussian process modeling
Abstract: The Global Historical Climatology Network-Daily database contains, among other variables, daily maximum and minimum temperatures from weather stations around the globe. It is long known that climatological summary statistics based on daily temperature minima and maxima will not be accurate, if the bias due to the time at which the observations were collected is not accounted for. Despite some previous work, to our knowledge, there does not exist a satisfactory solution to this important problem. In this talk, we will detail the problem and develop a novel approach to address it. Our idea is to impute the hourly temperatures at the location of the measurements by borrowing information from the nearby stations that record hourly temperatures, which then can be used to create accurate summaries of temperature extremes. The key difficulty is that these imputations of the temperature curves must satisfy the constraint of falling between the observed daily minima and maxima, and attaining those values at least once in a twenty-four hour period. We develop a spatiotemporal Gaussian process model for imputing the hourly measurements from the nearby stations, and then develop a novel and easy to implement Markov Chain Monte Carlo technique to sample from the posterior distribution satisfying the above constraints.