A1227
Title: Identifying shocks in oil futures returns curves using AR-MIDAS regression models
Authors: Meysam Sojoudi - University of Reading (United Kingdom) [presenting]
Abstract: The purpose is to pinpoint structural changes in the dynamics of oil returns and the forces behind those changes. The structural breaks test is applied in the AR-MIDAS regression models framework, an effective tool for modelling variables at various data frequencies. The Cross-Entropy algorithm, a quick and precise algorithm to detect structural breaks, is the foundation of our test. The empirical research, based on the pricing of West Texas Intermediate (WTI) crude oil futures, shows that there have been multiple shocks in the oil returns dynamics, including significant breaks in the mean and volatility of returns. The effects of a number of variables are analyzed, including the dollar index and energy indices, as well as events that could change the oil returns dynamic.
