Dynamic structure of the spot price of crude oil: does time aggregation matter?

Abstract

This paper assess nonlinear structures in the time series data generating mechanism of crude oil prices. We apply well-known univariate tests for nonlinearity, with distinct power functions over alternatives, but with different null hypotheses reflecting the existence of different concepts of linearity and nonlinearity in the time series literature. We utilize daily data on crude oil spot prices for over 26 years, as well as monthly data on crude oil spot prices for 41 years. Investigating the monthly price process of crude oil distinguishes this paper from existing studies of the time series structure of energy markets. All the tests detect strong evidence of general nonlinear serial dependence, as well as nonlinearity in the mean, variance, and skewness functions in the daily spot price process of crude oil. Since evidence of nonlinear dependence is less dramatic in monthly observations, nonlinear serial dependence is moderated by time aggregation in crude oil prices.

Nonlinearity in Global Crude Oil Benchmarks: Disentangling the Effect of Time Aggregation

“…Following the precedence in literature (Aghababa & Barnett, 2016; Lim, 2009; Lim & Brooks, 2009; Madhavan, 2014; Varghese & Madhavan, 2019), we employ the nonlinearity toolkit proposed by Patterson and Ashley (2000). To be specific, we employ a battery of nonlinearity tests, namely BDS test, Engle’s lagrange multiplier test, Hinich bicorrelation test, Hinich bispectrum test, McLeod–Li test, and Tsay test, so as to test for nonlinear serial dependencies in Brent, Dubai and WTI crude oil markets.…”

“…The distinct role that the benchmark crude return dynamics play in the global economic performance, as well as that of financial markets world over, has motivated scholars and practitioners to increasingly focus on crude oil price dynamics over the past few decades. Keeping in view the impact that several of the past socioeconomic and political events have had on determining crude oil prices, the recent studies on crude markets pinpoint to the prevalence of nonlinear serial dependence in the underlying data generation processes (Aghababa & Barnett, 2016; Haidar & Wolff, 2011). Using multiple tests for nonlinearity on daily spot prices, Kyrtsou et al (2009) found crude oil price fundamentals to significantly drive the market prices along with that of exogenous shocks.…”

Long memory dynamics and relative efficiency of crude oil benchmarks: An adaptive market hypothesis perspective

“…In other words, besides macroeconomic factors that affect all financial markets in general, several market‐specific factors also exert substantial influence on the data generating process of crude oil markets. For instance, unforeseen events such as oil embargoes, wars, market deregulations, and revolutions among many other geopolitical events have historically altered the growth trajectory of evolving crude oil prices (Aboura & Chevallier, 2016; Aghababa & Barnett, 2016; Lescaroux & Mignon, 2008; Varghese, 2017;Varghese & Raju, 2020). Consequently, it is only reasonable to assume the prevalence of multiple structural breaks throughout the sample period.…”

A general property for time aggregation

“…But non-linear models are more difficult to handle and very few attempts to derive aggregation results have been successful. 2 Much of the data used in finance and economics shows strong evidence of non-linearity in its structure, so theoretical results on temporal aggregation for non-linear models are particularly important for such applications [Aghababa and Barnett, 2016].…”