Dan J. Laughhunn

Quadratic Binary Programming with Application to Capital-Budgeting Problems

The purpose of this paper is to present an algorithm for solving the quadratic binary programming problem. Although a problem with this structure may arise in many situations, it is particularly common in capital budgeting when a decision-maker is confronted with a set of investment proposals from which he must select a portfolio. If returns of proposals are intercorrelated random variables and if the decision-maker uses as his criterion for selection the mean μ and variance σ2 of portfolio returns, his decision requires prior identification of the (μ, σ2) efficient set. The algorithm developed to solve the problem and hence necessary to generate the efficient set is based on the concept of implicit enumeration recently introduced by Egon Balas for solution of the binary linear programming problem.

The Impact of Sunk Outcomes on Risky Choice Behavior.

AbstractThis paper examines the impact of historical or stmk outcomes on the framing of risky choice problems. Two conflicting views exist in the literature about framitig stich sunk outcomes: framing based on the minimal account (where sunk outcotnes arc ignored) and framing based on the psychological account (where sunk outcomes are somehow included in the decision process). Several studies are reviewed in order to identify the ratiotiale for the two cotiflicting views of framing and to isolate some of the factors that are of potential importance in determining the appropriate decision frame of sunk outcome problems. Results of an exploratory experiment on framing of risky choice problems with sixty professional managers arc also reported. In general, these results suggest a rather extensive use of the psychological account for both personal and corporate decisions, for continue/ discontinue problems and asset disposal problems, and for sunk betiefits as well as sunk costs. Experimental results also ind...

Risk Preference: Empirical Evidence and Its Implications for Capital Budgeting

Finance is a subject that must inevitably deal with decisions that involve choices from risky alternatives in a wide variety of settings—for example, choices of investment portfolios or selections of capital assets. In order to develop normative and predictive models for choice problems such as these, assumptions about the risk preference of individual decision makers have been necessary. The traditional assumption made about risk preference is that individuals are uniformly risk averse. In its strongest form, the assumption of risk aversion has been translated into the proposition that individuals choose between risky alternatives on the basis of mean and variance (or semivariance) and that individuals are averse to risk as measured by variance (or semivariance). In a weaker form, the assumption of risk aversion implies that the utility function of individuals is concave in terminal wealth.

Risk attitudes in the telecommunications industry

It has been suggested in the economics literature that strongly risk-averse individuals may tend to gravitate toward large monopolistic firms rather than work for more competitive companies. In this article we provide results of an experiment designed to test this hypothesis. Operating managers from both regulated telecommunications firms and a broad cross section of unregulated industrial and service corporations were placed in the same controlled decision context to examine whether there are any systematic differences between telecommunications managers and their counterparts in nonregulated companies in their risk attitude toward losses.

Distributional Effects in Demand Analysis: Observations and Predictive Tests

Abstract Most previous empirical demand analyses using time series data have ignored the possible demand effect due to temporal variation in the populations demographic structure. Following an earlier study by Lippitt, this demand effect is denoted as a distributional effect. This article summarizes results of predictive tests designed to isolate the contribution to predictive accuracy attributable to the distributional effect. Nine expenditure categories are included in the predictive tests. With the exception of two expenditure categories (house operation and house furnishing), the distributional effect was not found to contribute to predictive accuracy.

On the Optimality of Single-Item, Incremental Cost Rules for the Make-Buy Decision

ABSTRACT Decision rules for the make-buy decision most often assume the form of single-item, incremental cost rules which, for application, require decisions to be made for each item in isolation from all others. While it has been recognized that such decision rules may be non-optimal in situations where equipment capacity limits exist, the literature has not considered other cases where non-optimality arises. The purpose of this paper is to use linear programming as a framework for evaluating the optimality of single-item, incremental cost rules. The major conclusion of this paper is that the presence of any type of binding constraint in the planning problem, regardless of its origin, renders a single-item decision rule potentially non-optimal. A set of optimal screening decision rules is developed to be used in conjunction with a linear programming model for make-buy decisions in cases where such binding constraints exist.