Time-varying Yield Distributions and the U.S. Crop Insurance Program

The objective of this study is to evaluate and model the yield risk associatedwith major agricultural commodities in the U.S. We are particularly concernedwith the nonstationary nature of the yield distribution, which primarily arisesbecause of technological progress and changing environmental conditions. Preciserisk assessment depends on the accuracy of modeling this distribution. This problembecomes more challenging as the yield distribution changes over time, a conditionthat holds for nearly all major crops. A common approach to this problemis based on a two-stage method in which the yield is first detrended and thenthe estimated residuals are treated as observed data and modeled using variousparametric or nonparametric methods. We propose an alternative parametricmodel that allows the moments of the yield distributions to change with time.Several model selection techniques suggest that the proposed time-varying modeloutperforms more conventional models in terms of in-sample goodness-of-fit, out-of-sample predictive power and the prediction accuracy of insurance premium rates.