Topics in stochastic dominance : theory and application

The dissertation investigates some important aspects of managerial decision making under conditions of uncertainty. In the last three decades two prominent general approaches have evolved to deal explicitly with risk in managerial decisions. They are: (1) the central tendency-dispersion trade off approach, and (2) expected utility analysis. The first task undertaken in this investigation is to integrate these two approaches. This is accomplished by identifying those situations in which decision rules obtained by either approach are equivalent. Once equivalence between the two basic approaches to decision making under uncertainty is established, the focus shifts to the extension of these decision theories into situations involving multi-attribute outcome spaces. In particular, stochastic dominance rules for multivariate outcome distributions are developed. Two applications of stochastic dominance criteria are then presented, illustrating the relevance of the approach to theory development and management of resource systems. The first illustration demonstrates the application of stochastic dominance to portfolio diversification problems. Several results are obtained describing the sensitivity of optimal mixes with respect to changes in opportunities for investment. The second illustration demonstrates the role stochastic dominance criteria can play in ecosystem policy analysis. A methodology of stochastic dominance policy screening for forest management systems is developed and applied.Business, Sauder School ofGraduat