Statistics Seminar: STA 290
Thursday, February 2nd, 2012 at 4.10pm, MSB 1147 (Colloquium Room)
Refreshments: 3.30pm, MSB 4110 (Statistics Lounge)
Speaker: Hao Zhang (Dept. Statistics, Purdue University)
Title: Infill Asymptotics in Spatial Statistics
Abstract: I will introduce some recent results under the infill asymptotic framework in spatial statistics and show how the results can be used to reduce computation for estimation and prediction for massive spatial data. Under the infill asymptotic framework, all data are sampled from a bounded region. Not all parameters are consistently estimable under this asymptotic framework. I will show how this inconsistency can be used to benefit inferences. Finally, I will compare the infill asymptotic framework with the increasing domain asymptotic framework and provide some guidelines on which asymptotic framework shall be applied for a given finite sample.