Electronic Theses and Dissertations for Graduate School
Add My Work
Author Last Name
Bayesian Semiparametric Covariance Function Estimation for Stationary Gaussian Processes by Overwhelming Force
Restricted (Penn State Only)
Ensley, John Robert
Master of Science
Date of Defense:
November 16, 2017
Benjamin Adam Shaby, Thesis Advisor
David Russell Hunter, Committee Member
Ephraim Mont Hanks, Committee Member
monte carlo integration
We use a fully Bayesian framework to semiparametrically estimate covariance functions of stationary Gaussian processes from irregularly-spaced data. The covariance functions are modeled in the spectral domain on the log-log scale using penalized natural splines, which results in a flexible class of smooth densities that have power law tails. The Fourier transforms required to construct likelihoods from the spectral densities are computed using standard Monte Carlo integration. These calculations require massive computational effort, but are highly parallelizable. By utilizing GPUs, computational speeds are competitive with standard parametric covariance model estimation.
Login using your Penn State access account to view the paper.