ASSET PRICING THEORY AND THE PREDICTABLE VARIATION IN AGRICULTURAL ASSET RETURNS

This study focuses on the time-series predictability of agricultural returns and evaluates the ability of conditional asset pricing models to capture such predictability. The models all have time-varying conditional risk premia and constant betas, as opposed to the unconditional specifications in prior studies. Results indicate significant time-varying predictability in agricultural asset returns. An Arbitrage Pricing model with prespecified economic factors, including a market portfolio, shows some ability to explain the predictability. The greater flexibility and information in conditional asset pricing models make them more suitable than unconditional models for many problems in agricultural finance.