Some Problems in Spatial Modeling: Simulation and Multivariate Random Fields

Abstract

Simulation of random fields is a fundamental requirement for most spatial analyses. For small spatial networks, simulations can be produced using direct manipulations of the covariance matrix. Larger high resolution simulations are most easily available for stationary processes, where algorithms such as circulant embedding can be used to simulate a process at millions of locations. We review some classic approaches for simulation of stationary random fields, and discuss how some of these can be extended to the nonstationary setting. The second half of the talk will cover some basic ideas for stochastic modeling of multivariate fields, and will discuss the connections between cross-covariance and spectral coherence.

WILL KLEIBER

University of Colorado Boulder