Date of Award
2026-05-01
Degree Name
Master of Science
Department
Mathematical Sciences
Advisor(s)
Rene Gutierrez Marquez
Abstract
Neural networks have demonstrated remarkable predictive performance in complex, high-dimensional settings. However, their highly parameterized structure and nonlinear architecture often make them difficult to interpret, limiting formal statistical inference and principled uncertainty quantification. In contrast, Gaussian processes (GPs) provide a fully probabilistic framework with a relatively small number of hyperparameters, enabling coherent uncertainty quantification and seamless integration into hierarchical and structured statistical models. Despite their theoretical flexibility as nonparametric function approximators, standard GP formulations frequently exhibit weaker predictive performance than modern neural networks in large-scale applications.
In this work, we employ a Gaussian process covariance function designed to approximate the functional behavior of neural networks. By embedding this construction within a statistical modeling framework, we bridge the predictive strength of neural architectures with the inferential advantages of Gaussian processes, yielding a model that supports both competitive predictive ability and rigorous uncertainty quantification. We apply the proposed methodology to Core-Based Statistical Area (CBSA) level housing price data at monthly resolution, modeling the month-over-month percentage change in median home price, illustrating its effectiveness for statistical inference and uncertainty assessment in spatially structured economic outcomes.
Language
en
Provenance
Received from ProQuest
Copyright Date
2026-05
File Size
74 p.
File Format
application/pdf
Rights Holder
Bismark Nyarko
Recommended Citation
Nyarko, Bismark, "Deep Neural Network-Gaussian Process for Housing Price Analysis" (2026). Open Access Theses & Dissertations. 4750.
https://scholarworks.utep.edu/open_etd/4750