Sequential binary investment decisions. A Bayesian approach (Q1187682)

This book is a revised and translated version of the author's `Habilitationsschrift', which has been accepted by the faculty of Economics and Social Sciences at the University of Graz, Austria. Dynamic models of investment theory are described. In the framework of dynamic Bayesian decision models, the formulation and analysis of investment processeses are concentrated to binary dynamic decision models. The book is devided into four chapters: Chap. 1 yields basic concepts of uncertainty and risk aversion. Chap. 2 gives an overview of the statistical results which are used in the subsequent sections (Monotone transition probabilities). Chap. 3 deals with the analysis of two-asset dynamic portfolio models. In a first part, the optimal investment policy and its properties are discussed in the framework of the classic dynamic portfolio model. Second, a stochastic dynamic approach for the portfolio model is presented, with stopping rule and monotonicity properties. Sensitivity analyses are based on an algorithm for two-point return distributions with variations of the utility function, prior distribution, and planning horizon. In chap. 4, a reversible real investment project with random net present value is discussed. A model for the determination of the optimal investment date under conditions of uncertainty is formulated as a stopping decision model. Sensitivity investigations with respect to capital costs of the project, planning horizon, discount factor, and prior distributions, conclude the approaches of optimal timing of investments.