Irreversible decisions under uncertainty. Optimal stopping made easy (Q2383143)

In this book an alternative aproach to optimal stopping problems is presented. Its basic ideas and techniques are demonstrated on a much simpler level than in the standard literature on optimal stopping theory. Necessary notations and results of the theory of stochastic processes are introduced in a piece meal basis when they are needed. Certainly, the reading of the book is easier for readers familiar with the theory of stochastic processes. The first three parts of the book can be used as an introductory course, and the rest of the book for an advanced course. The book has five parts. Part I regards time-discrete space models with finite time horizon. Part II extends part I to infinite time horizon. Part III treats discrete-time continuous space models. Part IV extends Part III to continuous time. Part V closes the book with extensions. Part I has three chapter. Chapter 1 (Introduction) treats uncertainty and (partial) irreversibility, option value of waiting, bad and good news principles, optimal stopping and stochastic control, discounted utility anomalies, models of uncertainty, choice of probability measure, techniques of this book, and gives an overview. Chapter 2 (Real options and American options) contains several examples illustrating the basic ideas and types of options in the simplest set-up of two-period and three-period models. Chapter 3 (Risk neutral pricing; finite time horizon) gives a short overview of the main definitions and results of the financial economic in the set- up of finite-state-finite time horizon models. Models proposed in chapter 2 and 3 can be solved by backward induction. In part II, the underlying stochastic factor is a random walk on an integer lattice and the time horizon is infinite. Backward induction is not longer applicable. In this case the Bellman equation is used as the Wiener-Hopf factorization for the latter. So, chapter 4 regards random walk on \(\mathbb Z\). Chapter 5 treats options in the binominal and trinominal models. Chapter 6 introduces the general random walk on \(\mathbb Z\) and option pricing. In part III additional problems are treated as investment lags, the expected waiting time for investment, capital dynamics. Certain advantages of the the discrete time modeling against continuous time modeling are demonstrated. So, chapter 7 regards random walks on \(\mathbb R\), chapter 8 treats basic options, and chapter 9 investigates optimal stopping for general random walks. In part IV, the investment threshold for the capital expansion program under non-standard specifications of the demand stocks are calculated. It is shown that the monopolist may find it optimal to increase the production and, simultaneously, to decrease the price. A model of technology adoption is considered. One factor is the price uncertainty, and the other one describes the evolution of the frontier technology. So, chapter 10 treats the Brownian motion, chapter 11 the general Lévy process, and chapter 12 embedded options. Part V derives an explicit backward procedure for pricing American options in continuous time models with finite time horizon. Then, for processes with jumps, the general exercise boundary is separated from the strike by a nonzero margin. The general option exercise rules can be used as a rule of thumb for more involved models of uncertainty. The standard problems in the model with mean-reversion are solved. For the standard perpetual American put and call options, the general formulas coincide with the exact results, and in exit and entry problems, the option values obtained from the general formulas differ from the exact one by no more than 3--5\% unless the mean reversion coefficient is large. So, chapter 13 treats American options with finite horizon. Chapter 14 investigates perpetual American options and real options under Ornstein-Uhlenbeck processes.