Search

It is widely assumed in portfolio theory that inveher hand, the main empirical foundation for such behavior in portfolio selection appears to be the observation of diversification. Risk-averse, expected-utility maximization implies diversification in portfolio selection, and investors are observed to diversify. In recent papers it has been argued that investors in some situations appear to display risk aversion to opportunity loss rather than to outcome. Thus I have argued [3] that investors are frequently jittery with very large cash positions as well as with very small ones. A natural explanation is that they fear being caught with a large cash position, if the market should rise, just as they fear being caught with a large stock position, if the market should fall. Being subject to a large opportunity loss of either type is cause for concern. The risk-averse expected-utility maximizer, on the other hand, presumably feels quite secure when he has little doubt about the outcome as he does with a large cash position. This is not the only evidence consistent with risk aversion to opportunity loss: I have argued [2] that the familiar strategy of dollar averaging is designed to hedge against large opportunity losses rather than against outcome variability. In particutar, under conditions that may be taken to approximate those under which dollar averaging is advocated, it has been shown that dollar averaging is a nonsequential, minimax strategy and cannot be a nonsequential, risk-averse, expected-utility maximizing strategy. Under similar conditions, Bawa [1] has recently pointed out that a sequential rather than nonsectuential minimax policy can also be interpreted in terms of a familiar prescription for stock-market trading. That prescription is to let your profits run but to cut your losses. [ABSTRACT FROM AUTHOR]

A model for minimizing the average cash balance subject to a constraint on the probability of stock-out is presented. The cash balance is described as an inventory process that changes because of deterministic and stochastic events. Recursive systems of equations are given to find (1) the distribution function of the cash level at any time and (2) the probability that all demands during some time interval are met. Then we examine the problem: minimize the expected time average cash balance subject to the condition that the probability that all demands are satisfied is at least some given number. It is shown that the optimal policy has a very simple form, which can be expressed verbally as, 'never have any more cash on hand than is necessary to satisfy the constraint.' [ABSTRACT FROM AUTHOR]

This paper shows that the widely appearing "marginal cost" formulations of the dynamic, periodic review, stochastic cash balance problem are equivalent to a "total cost" formulation, as long as the holding, shortage cost, and "closing of account" functions are constructed correctly. [ABSTRACT FROM AUTHOR]

A lack of synchronization between cash inflows and outflows presents the problem of investing positive net flows and financing negative net flows. In addition to the basic investment-financing issue there are other interrelated problems such as the payment schedule and the minimal cash balance that directly affect cash management decisions. Previous attempts to solve these problems have imposed constraints that limit their application. The purpose of this study is to develop an applicable method for solving the cash management problem by deriving a linear programming model that is divided into unequal periods. The unequal time division maintains the day-to-day aspect of the problem while reducing the number of periods and variables to a manageable size. Decision variables are divided into four groups: the payment schedule, securities transactions, short-term financing, and the cash balance. The model determines the optimal values of these variables subject to institutional constraints and within a framework of managerial risk preferences that are expressed as subjective constraints and subsequently evaluated by the dual model. Although a certainty assumption is made, inputs are analyzed by sensitivity analysis and the dual model, while the effects of uncertainty are further reduced by the unequal time structure of the model. Tests of the model indicate that it is both profitable and generally applicable by most business firms. [ABSTRACT FROM AUTHOR]

The problem considered in this paper deals with the optimal policies for keeping cash in anticipation of future net expenses. It is an inventory problem in which the inventory level can either decrease or increase, and the decision maker is allowed to change the inventory level in any direction at the beginning of each period. In addition to the usual holding and shortage costs, the problem involves fixed and proportional costs for deciding to change the inventory level. This paper shows the form of the optimal policy if the expected holding and shortage cost is convex and if deciding either to increase or to decrease the inventory level does not involve a fixed cost. [ABSTRACT FROM AUTHOR]

The paper describes a very apparent application of heuristic problem solving to accounts receivable management, namely in the location of lock-boxes. Lock-boxes are post office boxes selected in a company's distribution area in order to minimize the time necessary for the company to collect and to have available the funds remitted by its customers in payment of their obligations. The paper starts with a statement of and solution to the problem of measuring a given set of locations. Next the difficulties with the formulation are pointed out and a heuristic program to reach a satisfactory solution is described. The program is written in FORTRAN IV and is currently being tested. [ABSTRACT FROM AUTHOR]