Energy Price Forecasts and the Effect of Oil Price Shocks on the Stock Market in Selected Countries

Abstract

In this dissertation, I construct different oil prices forecasting models and estimate them with 128 monthly indicators that tell a complex macroeconomic story of the US economy from 19592020. This dissertation deals with presenting and estimating different oil price forecasting models and studying the relationship between stock market returns in some countries and the oil price changes. The study divides the forecasting models into linear and nonlinear models. In linear models, we consider some traditional time series models, such as random walk, autoregressive (AR), moving average (MA), and autoregressive integrated moving average (ARIMA) to forecast oil prices. The dissertation considers using principal component analysis (PCA) and partial least squares (PLS) to extract factors to be used in addition to the ARIMA process (ARIMAX).