Hedging downside risk in agricultural commodities: A novel nonparametric kernel method

Abstract

Using a nonparametric kernel method, this paper develops a weighted conditional value-at-risk hedge model to hedge downside risks in agricultural commodities. The model exhibits convexity, ensuring the acquisition of its global optimal solution. Simulations show that the nonparametric kernel method enhances the accuracy of the weighted conditional value-at-risk and hedge ratio determination, outperforming traditional estimation methods. Using major agricultural commodities, empirical evidence shows the superiority of the proposed model in reducing downside risks, compared to the minimum variance, minimum value-at-risk, and minimum conditional value-at-risk hedge models.