Exploring macroeconomic impacts on agricultural spot markets with time series methods

The increasing complexity and volatility present in agricultural commodity markets creates a need for the exploration of macroeconomic impacts on markets in the United States. This study uses two time series econometric methods to explore the possible impacts on agricultural spot markets. The first essay analyzes the recent commodity price run-up by investigating the dynamic correlations between corn, wheat, oil, exchange rates, and gold. By analyzing changes in the daily price series innovations, the research looks to test the hypothesis put forth by various authors for the reasons associated with the run-up. The second essay uses a latent variable modeling technique to analyze the implicit impact of macroeconomic factors on the beef sector. By uncovering the validity of the model for forecasting, the latent macroeconomic variables used for the forecasts will provide some understanding of how macroeconomics impacts the sector and lend themselves to increasing the forecasting. Results from the first essay show time-varying correlations between corn, oil, and exchange rates are present in the data and for the most part closely follows the suppositions put forth in the literature. However, correlations between wheat, oil, and exchange rates do not provide the clear picture present in the corn estimations. Results from the second essay show some capability for macroeconomic factors to improve forecasting performance. While the span and time frame for improved forecasting is short, the ability to improve forecasts in the sector for a month is significant.