Skip to Main Content
 

Global Search Box

 
 
 

ETD Abstract Container

Abstract Header

Essays on Risk Management for Agricultural Commodity Futures Market

Wang, Ying
http://rave.ohiolink.edu/etdc/view?acc_num=osu1461192690

Abstract Details

2016, Doctor of Philosophy, Ohio State University, Agricultural, Environmental and Developmental Economics.
Funding risk, which caused the $1.3 billion derivatives-related loss at MG Refining & Marketing, Inc. in 1993, has long been overlooked in the risk management literature. The key to understanding funding risk is that, as futures hedging practice requires substantial infusions of cash to meet variation margin calls, the maximum margin required may occur well before the expiration of the futures contract, but must be met in order to maintain the futures positions. This paper approaches the question of how to properly measure the “funding risk” of commodity futures positions by estimating a CD-Vine copula model for the dependence of corn, soybean and wheat futures at multiple forecast intervals, using Harrison’s method and the Extreme Value Theory to calibrate the distribution of the maximum. This is the first attempt in the literature to model the extreme prices of futures contracts over a given time period in an agricultural commodity portfolio context. The adoption of the recently-developed CD-Vine copula model allows one to model the dependence structure in a more flexible manner than the previous standard multivariate models based on Gaussian or Student’s t distributions. Witnessing the recent surge in price and volatility of agricultural commodity markets, it cannot be emphasized enough how important it is to assess the probability of rare and extreme price movements in the risk management of agricultural commodity futures. Similar to other financial time series, agricultural commodity futures exhibit the characteristics of time-varying volatility and fat tails. In this chapter, we employ the McNeil and Frey’s two step approach and conditional Extreme Value Theory to estimate Value-at-Risk (VaR) and Expected Shortfall (ES) for long and short positions in the agricultural commodity futures market at multiple significance levels, and compare this approach to conventional multivariate Normal or Student’s t distribution based models, Historical Simulation, RiskMetrics etc. The backtesting demonstrates that this GARCH-EVT approach provides a significant improvement over the widely used Normal and Student’s t distribution based VaR and ES models, which tend to underestimate the true risk and fail to provide accurate VaR estimates that are statistically no different from the corresponding significance level. To capture the tail dependence and properly estimate portfolio VaR, copula models are introduced to estimate a portfolio measure of risk in a multi-commodity setting. This has broad applications, for instance, for an agricultural commodity end-user that is purchasing corn, wheat and soybeans simultaneously. The conventional approach to this problem is to use a multivariate GARCH (a.k.a. MGARCH) model to estimate the conditional covariance between the futures prices. However, the typical MGARCH model approach inevitably suffers from being unduly restrictive because of the classical joint multivariate Gaussian assumption, despite the empirical evidence against elliptical distributions in commodity price returns. Also, from the perspective of computational efforts, the number of parameters to be estimated in the MGARCH specification often increases rapidly, stemming from the high-dimensional nature of the problem.
Matthew Roberts (Advisor)
Stanley Thompson (Committee Member)
Mario Miranda (Committee Member)
132 p.
Agricultural Economics; Economics

Recommended Citations

Citations

  • Wang, Y. (2016). Essays on Risk Management for Agricultural Commodity Futures Market [Doctoral dissertation, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1461192690

    APA Style (7th edition)

  • Wang, Ying. Essays on Risk Management for Agricultural Commodity Futures Market. 2016. Ohio State University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1461192690.

    MLA Style (8th edition)

  • Wang, Ying. "Essays on Risk Management for Agricultural Commodity Futures Market." Doctoral dissertation, Ohio State University, 2016. http://rave.ohiolink.edu/etdc/view?acc_num=osu1461192690

    Chicago Manual of Style (17th edition)

osu1461192690
956
© 2016, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.