K.C.S.Pillai Memorial Lecture
Statistical models for spatial processes on a global scale
Michael Stein
Professor
Department of Statistics, University of Chicago
Start Date and Time: Thu, 21 Sep 2006, 4:30 PM
End Date and Time: Thu, 21 Sep 2006, 5:30 PM
Venue: MATH 175
Refreshments: Refreshments will be served in the Math Library at 4:00 p.m.
Abstract:
Statistical models for atmospheric processes on a global scale need to respect the Earth's spherical shape and the common property of processes behaving fundamentally differently as latitude varies. Using total column ozone as an example, this talk explores these issues and considers various ways of developing global models that capture the observed covariance structure of total column ozone. The emphasis will be on truncated series expansions in spherical harmonics, which provides a natural way of generating models that are invariant to rotations about the Earth's axis and make good likelihood approximations using the massive dataset (over a million obsevations a week) feasible.