Multiproduct Nonconstant and Time -Varying Hedge Ratio Estimation by Locally Polynomial Kernel Applied to the Hog Complex

Multiproduct hedge ratios generated by locally polynomial kernel are recommended to use when the futures market is utilized to manage price risks because they are found to take account of more information inflow on prices than GARCH-type models. Estimated hedge ratios using past information might not be relevant to hog producers to manage their price risks in reality. Thus, hedge ratios are one-step ahead forecasted, ex ante hedge ratios, and evaluated for an out-of-sample period. Multiproduct hedging using LPK is suggested for use when the futures market is needed to hedge price risks in a forthcoming period, which gives risk minimizing hedge ratios with higher mean return than other hedging strategies. In addition, this study evaluates the selective hedging using forecasted prices by LPK, which has smaller forecasting errors than the vector autoregression model, and found that the unhedged variance of return is substantially reduced by selective hedging, but not reduced as much as by full hedging strategies. Meanwhile, the return by selective hedging increases considerably relative to full hedging. Thus, the selective hedging is a good alternative to hog producers who do not want to sacrifice mean return as much as when full hedging is used, and want to retain a lower variance of return.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Agricultural Economics

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Kim, Min-Kyoung

Contributors dc:contributor

• Leuthold, Raymond M.

• Economics, Finance

Language dc:language

eng

Identifier

(MiAaPQ)AAI3017121

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/82941