Agric. Econ. - Czech, 2021, 67(8):305-315 | DOI: 10.17221/148/2021-AGRICECON

Measuring parametric and semiparametric downside risks of selected agricultural commoditiesOriginal Paper

Dejan Živkov^*,1, Marijana Joksimović², Suzana Balaban²

¹ Novi Sad School of Business, University of Novi Sad, Novi Sad, Serbia

² Faculty of Finances, Banking and Audit, Alfa University, Belgrade, Serbia

In this paper, we evaluate the downside risk of six major agricultural commodities - corn, wheat, soybeans, soybean meal, soybean oil and oats. For research purposes, we first use an optimal generalised autoregressive conditional heteroscedasticity (GARCH) model to create residuals, which we later use for measuring downside risks via parametric and semiparametric approaches. Modified value-at-risk (mVaR) and modified conditional value-at-risk (mCVaR) provide more accurate downside risk results than do ordinary value-at-risk (VaR) and conditional value-at-risk (CVaR). We report that soybean oil has the lowest mVaR and mCVaR because it has two very favourable features - skewness around zero and low kurtosis. The second-best commodity is soybeans. The worst-performing downside risk results are in wheat and oats, primarily because of their very high kurtosis values. On the basis of the results, we propose to investors and various agents involved with these agricultural assets that they reduce the risk of loss by combining these assets with other financial or commodity assets that have low risk.

Keywords: Cornish-Fisher expansion; generalised autoregressive conditional heteroscedasticity (GARCH) model; grains

Published: August 26, 2021  Show citation