DECISION CRITERIA IN RISK ANALYSIS: AN APPLICATION OF STOCHASTIC DOMINANCE WITH RESPECT TO A FUNCTION

Decision criteria under risk are briefly discussed and evaluated. It is concluded that stochastic dominance with respect to a function is more flexible and powerful than other decision rules. By placing limits on the decision makerʹs degree of absolute risk aversion, this method reduces a large number of actions to an efficient set that are not dominated for the specified class of decision makers. Thus, there is no need for detailed knowledge of the decision makerʹs preferences. Using utility efficient programming model, it was illustrated that this procedure can effectively incorporate the stochastic dominance with respect to a function criterion in whole-farm planning. Results of this study demonstrated that risk aversion plays an important role in farmersʹ behavior. Ignoring this factor can lead to selecting programs which overestimate farmersʹ levels of total net revenue. This may explain why results of deterministic models based on assumed risk neutrality are often disregarded by farmers as unrealistic.