Incorporating historical weather information in crop insurance rating

Crop insurance programs rely on conditional predictive distributions of loss random variables (e.g., yield, revenue, costs, etc.) to determine probabilities and magnitudes loss. The may be related stochastic that are not known at the time policy is priced. Such case for weather; weather stochastic, realizations when crop sold, there often additional historical information relative variable itself. We provide a Bayesian methodology incorporating in rating. apply method empirical applications county-level U.S. corn yields cost ratios Midwest. weather-conditioned differ from those based shorter samples. In yield distribution setting, temporal leads economic gains other rating approaches; magnitude these increases with amount included analysis.