Graduate Project

WTI Oil Price Prediction Modeling and Forecasting

This work examines two different Baysian approaches to model short term oil price return for past decades and forecast it. We frstbuilt the multivariable linear regression model based on relevantexplanatoryvariables. Then webuild the univariate time series model using ARIMA models, followed by ARCH and GARCH models. Both methods are followed by required procedures and econometrics tests. The forecasting powers of time series approach perform better than linear regression and even structural models, yet linear approachisveryrelevantforknowing incapabilityofeachvariabletooilprice. iv

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.