Prediction Methods for Semi-continuous Data with Applications in Climate Science
Thursday, April 27, 2017 at 4:00pm to 5:00pm
7800 York Road, Room 320 7800 York Rd, Towson University, Towson, MD 21204, USA
Soeaker: Dr. Nagaraj K. Neerchal,
The University of Maryland, Baltimore County.
Abstract: Semi-continuous random variables have discrete and continuous components with support on a set of discrete points and a subset on the real line. Daily precipitation (rainfall) data is an example of such a random variable with a point mass at 0 and an absolutely continuous distribution function on the positive real line. When the Probability of observing a 0 is assumed to be independent of the parameters of the continuous part, the density of the random variable takes the form a Two-Part model. A popular form that enforces a dependency is the standard Tobit model. We briefly review some inferential aspects of the semi-continuous distributions and present several methods of prediction and derivation of predictive densities, motivated by applications of spatio-temporal models in Climate Science.
(Joint Work with Sai Kumar Popuri, UMBC and Dr. Amita Mehta of Joint Center for Earth Systems Technology)
No recent activity