Essays on commodity investing and volatility risk

This dissertation consists of three essays that investigate issues in commodity investing and volatility risk in commodity futures markets. The first essay evaluates the usefulness of commodities in a portfolio by examining multiple commodity instrument tools and by controlling for estimation error. Using data from three generations of commodity indices and 15 individual commodity futures for 1991-2015, we find that including most commodities does little to improve the portfolio's Sharpe ratio especially in an out-of-sample context. The only exception is the third-generation commodity index that embeds a momentum strategy and can substantially enhance portfolio performance. When shrinkage estimators are used to reduce estimation errors in expected returns, the resulting portfolios are more diversified and stable over time, and more important, commodities play a much smaller role in terms of risk reduction in those portfolios. Our findings suggest that investors should be more cautious and selective in commodity investment activities. The second essay examines the impact of commodity index investment on WTI crude oil prices by focusing on mapping algorithms. Previous studies employ mapping algorithms to infer index positions in WTI crude oil from positions in agricultural commodities and find an economically large and statistically significant impact of index positions on oil prices. We provide direct evidence that the identified impact of index investment from mapping algorithms is highly questionable, with the entire forecasting power coming from 2008. More specifically, an idiosyncratic spike in agricultural index positions during 2008, coupled with the spike in oil prices, causes the spurious impact of index investment on crude oil futures prices found in earlier studies. The third essay investigates the pricing of volatility risk in agricultural commodity markets by examining delta-neutral straddle gains. Within a stochastic volatility model, delta-neutral straddle gains scaled by futures price are mainly determined by volatility risk premium and its risk exposure. Using a sample of options on corn, soybeans, wheat, live cattle, and lean hogs for 2003-2016, we nd that volatility risk is priced with a negative premium in agricultural commodity markets. Volatility risk premium shows a non-trivial term structure with its absolute value declining sharply in maturity and approaching zero beyond three months. Regression results reveal that volatility risk premium is related to expected volatility, time to maturity, and trading volume in futures, and becomes more evident on the day right before the release of USDA reports.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Agricultural & Applied Econ

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2018

Author dc:creator

• Yan, Lei

Contributors dc:contributor

• Garcia, Philip
• Irwin, Scott
• Sanders, Dwight
• Serra Devesa, Maria Teresa

• Commodity index
• Commodity futures
• Financialization
• Volatility risk

Statement dc:rights

• Copyright 2017 Lei Yan

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/99094

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/99094