Testing mean-reversion in agricultural commodity prices: Evidence from wavelet analysis
Vol. 12, No 4, 2019
Adedoyin Isola Lawal
Dept. of Accounting and Finance, Landmark University, Omu Aran, Nigeria lawal.adedoyin@lmu.edu.ng ORCID: 0000-0001-8295-1560 |
Testing mean-reversion in agricultural commodity prices: Evidence from wavelet analysis |
Oluwasola Emmanuel Omoju
National Institute for Legislative and Democratic Studies, Abuja, Nigeria solaomoju@gmail.com Abiola Ayopo Babajide
Dept. of Banking and Finance, Covenant University, Nigeria babajide.abiola@covenantuniversity.edu.ng Abiola John Asaleye
Dept. of Economics, Landmark University, Omu Aran, Nigeria Asaleye.abiola@lmu.edu.ng
|
Abstract. This study examines the validity of the random walk hypothesis for some selected soft agricultural commodity prices within the context of heterogeneous market hypothesis and mean reversion hypothesis. The study employs a battery of traditional unit root tests, GARCH-based models and a novel frequency-based wavelet analysis to analyze daily data sourced from 6th of Jan 1986 to 29th Dec 2018. Contrary to other existing studies that employed only traditional time domain unit root tests, our results reveal that soft commodity prices are mean reverting, suggesting the existence of potential excess returns for investors. Overall, our results show that the selected soft commodity series are inefficient when we factored in heteroscedascity and frequency domain into our model. Our study is an improvement on the existing studies as we analyze our data using both time and frequency domain estimates. Besides, unlike other studies that did not offer structural breaks, the current study provides structural break dates with major events in the global socioeconomic space, which are key to identifying the date of bubbles and potential signs of commodity price bubbles. Our findings have some critical implications for investors, policy makers, academics and other interested economic agents. |
Received: January, 2019 1st Revision: May, 2019 Accepted: November, 2019 |
|
DOI: 10.14254/2071-8330.2019/12-4/7
|
|
JEL Classification: J180, P5, D630, I38 |
Keywords: wavelet analysis, testing, policy makers |