Semester
Spring
Date of Graduation
2013
Document Type
Dissertation
Degree Type
PhD
College
Chambers College of Business and Economics
Department
Economics
Abstract
Nonlinear models have many applications in the economic and financial fields. The following works focus on their use for forecasting. Neural networks, in conjunction with an Affine Term Structure Model, are used to discover possible yield curve arbitrage opportunities. A G/ARCH model is employed to test and forecast the conditional variance of state-level employment growth. A Space-Time Autoregressive (STAR) model is applied to state employment growth to ensure that any measure of volatility in the series is not misdirected as employment movements between neighboring states.
Recommended Citation
Murdoch, Scott T., "Three Essays on Forecasting in Nonlinear Models" (2013). Graduate Theses, Dissertations, and Problem Reports (ETD). 531.
https://researchrepository.wvu.edu/etd/531