This dissertation consists of three papers that investigate different dimensions of commodity price variability which has increased dramatically in recent years. The first paper analyzes recent volatility spillovers in the United States from crude oil to corn and ethanol markets using futures prices. Spillovers to both corn and ethanol markets are somewhat similar in timing and magnitude, but moderately stronger to the ethanol market. The shares of corn and ethanol price variability directly attributed to volatility in the crude oil market are generally between 10%-20%, but reached nearly 45% during the financial crisis when world demand for oil changed dramatically. Volatility transmission is also found from the corn to the ethanol market, but not the opposite direction. The findings provide insights into the extent of volatility linkages among energy and agricultural markets in a period characterized by strong price variability and significant production of corn-based ethanol. The second paper investigates short-term price density forecasting procedures in the Lean Hog Futures Market. High price variability in agricultural commodities increases the importance of accurate forecasts. Density forecasts estimate the future probability distribution of a random variable, offering a complete description of risk. In this paper we develop short-term density forecasts of lean hog prices for the 2002-2012 period. For a two-week horizon, we estimate historical densities using GARCH models with different error distributions and generate forward-looking implied distributions, obtaining risk-neutral densities from the information contained in options prices. Real-world densities, which incorporate risk, are obtained by parametric and non parametric calibration of the risk-neutral densities. Then the predictive accuracy of the forecasts is evaluated. Goodness of fit and out-of-sample acii curacy comparisons indicate that real-world densities outperform risk-neutral and historical time series generated densities. This supports the notion that a risk premium exists even at a two-week horizon in the hog market and that market participants can use these forecast to develop a better understanding of the final distribution of prices. In the final paper, we develop and evaluate quarterly out-of-sample individual and composite density forecasts for U.S. hog prices. Individual forecasts are generated from time series models and the implied distribution of USDA, Iowa State University, and University of Missouri outlook forecasts. Composite density forecasts are constructed using linear and logarithmic combinations, and several weighting schemes. Density forecasts are evaluated on predictive accuracy and goodness of fit. Logarithmic combinations using equal and mean square error weights outperform all individual density forecasts and all linear combinations. Comparison of the outlook forecasts to the best logarithmic composite demonstrates the consistent superiority of the composite procedure, and identifies the potential to provide hog producers and market participants with accurate expected price probability distributions that can facilitate decision making.
|Qualification||Doctor of Philosophy|
|Award date||12 Feb 2013|
|Publication status||Published - 2013|