Search

The purpose of this paper is to reformulate the problem of optimization by firms facing stochastic demand in the context of the Sharpe-Lintner Capital Asset Pricing Model. We shall focus only on the perfectly competitive firm which produces a single, homogeneous output. Two conclusions are drawn. First, the existence of stochastic demand will not necessarily cause the perfectly competitive firm to alter its output from that which it would produce under certainty, even if the market is dominated by risk-averse individuals. The critical factor in determining any alteration in output due to uncertainty is the covariance between the price of the firm's product and the rate of return on a market asset portfolio. Second, and more important, our analysis implies that the cost of capital is not invariant to the levels of production of a firm. The risk shareholders confront is a function of the level of output, and two unlevered, competitive firms producing the same product may have different costs of capital. [ABSTRACT FROM AUTHOR]

The purpose of the Lessard study is to investigate the nature of the stochastic' dependence among stocks and relate such knowledge to both normative portfolio management and positive capital market theory. Specifically, it purports to investigate via multivariate analysis the implications of the dependence structure of security prices for:1. Capital asset pricing. 2. Normative investment decision making. Clearly, the paper is not theoretical in intent. Unfortunately, it is largely devoid of empiricism and particularly the type of empiricism to get much mileage out of the nature of the stochastic dependence among stocks. Indeed, knowledge of the nature of the stochastic dependence among stocks is not sufficient to discriminate between major competing capital asset pricing models or, for that matter, to shed light on any of the major controversies within this realm. [ABSTRACT FROM AUTHOR]

This paper examines the international structure of returns on common stocks with two specific purposes: to better understand the importance of international diversification relative to domestic diversification and to improve the specification of the stochastic process generating returns. [ABSTRACT FROM AUTHOR]

Strong evidence will be introduced which suggests that brand choice behavior is substantially stochastic. A general theory of stochastic preference is presented and tested. Brand switching data are shown to be in substantial agreement with the theory. [ABSTRACT FROM AUTHOR]

The objective of this paper is to determine upper bounds of the potential savings that can be realized by the application of cash management optimization models. These upper bounds are found by simulation as the difference between the performance of a deterministic optimization model--which finds the optimal policy in hindsight--and the simulated performance of a hypothetical treasurer who uses simple heuristic cash management rules as informally practiced by many treasurers, based on prediction of random cash flows. The results of this analysis leave serious doubts as to profitability of cash management optimization models. [ABSTRACT FROM AUTHOR]

A mathematical stochastic model for accounting evaluation of human assets¿synthesizing both the world views of economics and accounting and consistent with recommended accounting principles¿is proposed. [ABSTRACT FROM AUTHOR]

The article focuses on debate concerning the Phelps-Friedman "accelerationist" view of the Phillips curve. It states that empirical tests of Phelps-Friedman is suspect because the validity of the tests is predicated on the adequacy of a suspect hypothesis. It suggests that while it is reasonable to believe a sustained, constant inflation would be fully anticipated, this consideration is not helpful for the purposes of empirical estimation as inflation is never as sustained during periods of estimation as the mental experiment claims.

The article focuses on the specification of stochastic processes deriving financial asset behaviors, particularly outlining relationships between a stock's price, beta, and exchange listing. Observations are given showing the correlation between price per share and future returns of a given stock, regardless of risk values. The implications of such behavior on future beta change assessment are discussed. Empirical differences observed between equivalent risk stock returns on the American Stock Exchange and New York Stock Exchange are also highlighted.

The purpose of this paper is to study the problem of choosing an optimal monetary strategy within a stochastic control-theoretic framework. Assuming the sole objective of monetary policy is to stabilize the fluctuation of income about some desired value, the policymaker's optimal strategy consists of the optimal income-stabilizing responses of his control-variable or instrument, for instance, the monetary base, to estimates of conditions within his structural hypothesis. In the static, linear, Keynesian model of this paper it is assumed that the policymaker knows the multiplicative parameters with certainty and that only the additive terms are subject to stochastic error. In addition, it is assumed that the current values of these additive terms are not immediately observable and must be estimated by the use of contemporaneous survey data. The three additive terms are autonomous expenditures and the intercepts of the money supply and money demand functions. [ABSTRACT FROM AUTHOR]

A NOTABLE and provocative development in the recent literature has been the application of the theory of random walks to the analysis of price behavior in the stock and commodity futures markets. The basic hypothesis of the random walk theory is that a particular price series behaves as a simple stochastic process. Successive price changes are independent random variables, implying that the past history of a series generates no information that would be useful in predicting future price changes. It is the purpose of this paper to rectify this void in the literature by applying both a sophisticated statistical tool and mechanical filter tests to similar data. Spectral analysis along with several mechanical filters will be employed to test whether the live cattle futures prices move in a random or a systematic fashion. [ABSTRACT FROM AUTHOR]

The time series behavior of corporate incomes has generated considerable interest. The early work of Little suggests that successive changes in corporate incomes in the U.K. are independent. Subsequent empirical work suggests that the same conclusion applies to corporate incomes in the U.S. This article applies different methods from those of previous researchers, but the conclusion is unchanged. We conclude that, in general, measured annual accounting incomes for U.S. corporations follow either a submartingale or some very similar process. Because of small sample problems this conclusion is based on mean and median results; no attempt has been made to determine whether specific firms are systematic outliers. The conclusion that corporate incomes are submartingales has important implications for forecasters and researchers in accounting and finance. For example, it implies that attempts to smooth corporate incomes in the manner suggested in the accounting and finance literatures are not successful. It also affects the interpretation of the growth and decline of firms. [ABSTRACT FROM AUTHOR]

This paper proposes a Markov chain model to deal with the dynamic, multivariate, and probabilistic nature of a human need system. The proposed model offers a powerful tool to analyze the prevailing states and to predict the future state of the need system. [ABSTRACT FROM AUTHOR]

This article discusses a procedure for building a perceptual map of a market based on revealed preference data. Specifically, it suggests how the transition matrix, which is central to many stochastic models in marketing [2, 5, 7, 8, 14, 15], can be used as a similarity matrix for input to multidimensional scaling or cluster analysis in place of the more widely used judged similarity measures [3, 4, 18]. This article thus attempts to demonstrate a conceptual link between stochastic models of market share and individual choice models. Using brand-switching matrices as a basis for either perceptual mapping or market segmentation is clearly not without peril. For example, it is very difficult to decide which people are sufficiently homogeneous, or what time period is sufficiently stable, to use as data for building a transition matrix that is not seriously affected by an averaging fallacy. Furthermore, assuming a unique and constant ideal point seems to contradict the tendency of individuals to seek variety. Yet these problems exist in any application of perceptual mapping that is not individually based. This article has attempted to demonstrate that stochastic models and perceptual mapping are complementary rather than competing approaches to explaining choice behavior. Moreover, brand-switching data can serve as a useful input to either multidimensional sealing or cluster analysis. Whether brand-switching data prove to be a valuable similarity measure, however, can only be decided after extensive testing. [ABSTRACT FROM AUTHOR]

A measure of brand acceptance involves a refinement of the usual stochastic model market prediction. Using this measure, the effect of consumer promotions and purchasing habits on brand acceptance is investigated and a brand health indicator is developed. [ABSTRACT FROM AUTHOR]

This article explores the use of a brand choice stochastic model's mean value function in evaluating two models empirically, using a common set of purchase data. The linear learning model fit the data well, but its mean value function was not capable of making reasonable predictions of successive, aggregate purchasing statistics. Another brand choice model, the new trier model, was found to perform much better. The results suggest that model tests should not be restricted to the usual goodness-of-fit test, especially in situations of nonstationarity. A structural comparison of the two models focuses on their different approaches to nonstationarity. [ABSTRACT FROM AUTHOR]

Goodhardt and Ehrenberg have developed a useful stochastic model for analyzing period-to-period fluctuations in sales. In this article we generalize their model to allow for nonbuyers of the product category. A systematic bias in their simple negative binomial distribution (NBO) model is demonstrated. In fact if the proportion of nonbuyers is large, the simple model will be wrong. We also give explicit formulas and directions that allow a moderately sophisticated analyst to perform his own conditional trend analysis. [ABSTRACT FROM AUTHOR]

Little progress has been made in building realistic models of consumer's purchase behavior as a function of the dynamic environment in which she makes her buying decisions. This article analyzes, probabilistically, relationships among selected elements of a consumer's purchase decision: brand purchased, store visited, and size of purchase. [ABSTRACT FROM AUTHOR]

Many articles have discussed the value of interpreting preference data for a distribution of buyer preferences arrayed on a product attribute or dimension [1-3, 5, 6, 9]. Products are seen as ranging from light suds to heavy suds, weak chocolate flavor to strong chocolate flavor, etc. The premise is that differences in buyers' preference patterns reflect these underlying dimensions. This article's purpose is not to question the concept of this dimensional approach but rather to question the data collection procedure used to develop the preference distribution. The argument is that the paired comparison procedure, as currently used, can lead to misinterpretation of the underlying preference distribution. The key to developing a realistic preference distribution is to place each subject on the dimension such that his location represents his first-order preference. Consequently, the procedure used to collect such information should allow the investigator to identify each subjects preferences ordering of the dimension levels. There are many data collection procedures (paired comparison, ???, rank order, etc.) allowing the investigator to do this [3]. [ABSTRACT FROM AUTHOR]

With factor analysis as a method of estimating parameters, an empirical model of measuring brand loyalty for individual consumers based on frequency and pattern of purchases is presented. Despite some limitations, the method seems superior to stochastic models for generating robust measures at the individual level. [ABSTRACT FROM AUTHOR]

There has been considerable discussion of the validity of alternative measures of radio listening. Stochastic process models are used to explore the relationship between some common measures and, in the process, the sort of data that should ideally be collected is specified. [ABSTRACT FROM AUTHOR]

This article presents comments on A. S. C. Ehrenberg's article, "An Appraisal of Markov Brand-Switching Models," that raises questions about stochastic brand-switching models and Markov models in general. The authors discuss Markov vs. Vokram probabilities, whether Vokrams have managerial significance, Ehrenberg's test for stationarity, and the value of steady-state brand shares. The authors agree that Markov processes do not fill all brand-switching situations, yet they would not abandon Markov processes altogether, nor would they dismiss the Vokram approach.

A stochastic model that generates data for sales planning and control is described. An example is presented that shows how these data are used to plan short-run sales activities and train effective salesmen. In the model, changes in customer propensities to buy are treated as Markov processes. Finally, it is suggested that the concepts developed here may be computerized and integrated into existing systems for planning and control. [ABSTRACT FROM AUTHOR]

Using no mathematics other than a simple arithmetical example, thus article reviews fundamental difficulties in applying Markov theory to brand-switching data. No successful practical applications of the theory appear to be available. [ABSTRACT FROM AUTHOR]

This paper views total sales as a two-phase random phenomenon: a randomly determined number of customers, and a randomly determined dollar she of an individual sale. Some properties of total sales so viewed are derived and possible applications in a marketing context are indicated. [ABSTRACT FROM AUTHOR]

The present study relates to a transport organisation operating a fleet of aircraft and aims at determination of the optimal spare float of engines. The flow of the engines is viewed as a queueing problem in a closed loop system with variable demand and supply. The servicing facility consists of three stations in tandem. Two cases of the servicing facility operation are investigated, first when no inter-stage queue is permitted, and second when a queue length of one unit is allowed between the first and second stations. The solution is obtained by the Monte Carlo Simulation technique. The results are compared with those obtained from a reliability theoretical approach and conclusions are drawn. [ABSTRACT FROM AUTHOR]

A stochastic model has been proposed for variables estimation in auditing. This model can be used in conjunction with classical estimation techniques to predict what sample sizes will be necessary to obtain a desired standard error of estimate. The sample sizes will be a function of the auditor's estimates of the error rate and the first two momenta of the error distribution, as well as the distribution of book values which is known at the time of audit. The model can also be used with a Bayesian analysis, but more research is needed in this area before it is feasible to implement such an approach. The model focuses attention on the need to estimate two different parameters in an audited population-the error rate and the distribution of errors for items found to be in error. Techniques which fail to recognize this underlying structure (e.g., simple mean per unit estimation using sample values only) will probably not be of much value for auditors. [ABSTRACT FROM AUTHOR]

In this paper I have extended the results of the analysis originated in [9) on the optimal timing of control messages. The analysis centered on a controlled unit whose behavior could be described by a two-state Markov process. The model developed applies to a case in which a single or aggregated decision-making unit's performance is to be controlled by a single or aggregated goal (not necessarily constant over time). The model is a continuous-time finite-horizon model which makes use of a finite-state continuous-time Markov process and relies on continuous-time optimal control theory. The timing of a message is an attribute that affects the value of an information source. The timing issue also has important implications for the design of communication systems. For example, it is closely related to the timing of (and hence the devices used for) data collection, the data storage devices to be used, and the kinds of communications channels to be established. These latter factors obviously affect the cost of communication-information systems. [ABSTRACT FROM AUTHOR]

Recent theoretical and empirical work in portfolio theory has exhibited a natural evolution from the two-moment EV model popularized by Markowitz through the higher moment models to selection on the basis of the entire probability function. This latter approach, referred to as the Stochastic Dominance (SD) approach to portfolio selection, has been shown to be theoretically superior to all of the "moment methods" and has been the focus of an increasing volume of empirical work. In comparison with the EV model, however, the SD approach has one potentially significant deficiency from the standpoint of empirical testing. With a given set of portfolios to evaluate, the computer processor time required for the determination of the EV efficient set is quite short. One can calculate and sort means and variances on several hundred portfolios in a matter of seconds. In contrast, the processor time required for the determination of the SD efficient set for the same number of portfolios can be prohibitive. The purpose of this paper is to examine the factors responsible for the relatively time-consuming nature of empirical tests of SD efficiency and to present the major results of our search for efficient algorithms for conducting SD tests. Section I briefly examines the time-consuming nature of the tests for SD efficiency. Section II presents and evaluates; our most efficient algorithms for obtaining true SD efficiency results. Section lit examines an approach that yields approximate SD efficiency results with further reductions in required processor time. [ABSTRACT FROM AUTHOR]

The purpose of this article is to explore the relevance of the theory of Markov processes to the analysis of stock price movements. The present study was prompted by the work of Dryden [6], in which aggregate data on United Kingdom share prices were analyzed within a Markovian framework, and which indicated that it might be fruitful to apply the Markov model to more disaggregated data, specifically to individual stock price data. Markov theory is seen to be relevant to the analysis of stock prices in two ways: 1. As a useful tool for making probabilistic statements about future stock price levels. In this role it constitutes an alternative to the more traditional regression forecasting techniques to which it is, in many ways, superior. 2. As an extension of the random walk hypothesis. It will be shown that under a fairly general partial-adjustment model of stock price determination, price movements which do not display the random walk characteristic may be interpreted as following a Markov process. [ABSTRACT FROM AUTHOR]

This article examines the relationship between stock price behavior and changes in the capital structure of a firm. The authors examine how the random walk hypothesis has been applied to the behavior of common stock prices and discuss the argument that has grown over the variance of the price-change process. They believe that a resolution of this disagreement in necessary in order to develop a realistic valuation model for convertible securities. They also examine the interdependence among common stock prices and senior securities.

In many nonlinear programming applications the objective function has an inherent uncertainty that depends upon a set of random variables that have a known distribution. If one wishes to optimize the expectation of the objective, as suggested by the expected utility theorem, then as is shown here one can often solve such problems by modifying standard nonlinear programming algorithms. To illustrate what is involved, the details and justification for the application of the interior parametric sequential unconstrained maximization technique and the generalized programming method for the solution of such problems are given. Some related problems with stochastic constraints for which the solution method applies are mentioned and an example of a portfolio selection problem is given. [ABSTRACT FROM AUTHOR]

The article presents statistical analysis techniques to account for the stochastic changes of stock price series processes through averaging measures. The particular impact of averaging measures within the random walk model is discussed and circumventive methods are presented to avoid the generated errors.

The article focuses on the application of mathematical control theory to finance through a model of problems concerning simple dynamic cash balance. It comments on models of time dependent cash balances, solving them through a version of control theory maximum principle. It states that from the point of view of financial analysis, the solutions obtained were intuitively satisfying, and that the simple models needed to be extended to become more realistic and complicated. It suggests that further theoretical research in deterministic and stochastic control theory was required to expand on the models.

The results of our tests indicate that branching does tend to reduce variability within a metropolitan area. Given the data with which we worked, no conclusive statement can be made concerning the relationship of geographical dispersion to diversification of deposits. In this paper we have measured the impact of branching on deposit variability for one financial institution and for one type of deposit. Thus, our results should be viewed as a first step in studying this question. If data become available, further replication for different areas and different types of deposits, especially demand deposits, should be undertaken. [ABSTRACT FROM AUTHOR]

The article presents an analysis of the random walk theory which states that in a free competitive market, prices for goods and services should reflect all of the information available to participants in the market that influence its price. A statistical process for which the expected future value of a random variable is independent of past values of the variables is called a Martingale process. The random walk hypothesis requires that in a competitive market, price changes will result from Martingale processes. Under the random walk hypothesis, price changes could occur if the information that becomes available to market participants is random in its effect on market price, if all the market participants are told about all new information, and are told about any statistical dependencies in the information process.

The article focuses on a stochastic system's optimal design with dominating fixed costs. It states that the total costs for a given service level is the sum of that service level's fixed costs, the cost of failing to provide required services, and the variable costs of satisfying a demand. It mentions that components that make up fixed costs include capital costs associated with money tied up in operating reserves and the cost of labor required to operate the service centers. It states that it may be necessary to rely on value judgment for imputing the cost of failing to provide adequate service.

Using experimentally generated data and a method of estimating parameters developed by Massy, the linear learning model was tested as a predictor of brand repeat purchasing. The results differ from those reported earlier and indicate that the model may not be as useful as others have suggested. [ABSTRACT FROM AUTHOR]

The number of gate positions required at an airport, or the number of flights that can be accommodated at a given number of gate positions, depends on how efficiently each gate position is used. The airlines' schedule and the airport's operating policies influence the efficient use of the gate positions. A simple stochastic model describing the behavior of flights relative to their schedule was developed based on empirical information. The model was then used to study the influence of a common scheduling practice of bank operation on the requirements for gate positions. It was found that a completely uniform schedule generates the minimum requirement. A procedure is presented to estimate the number of gate positions required at an airport. [ABSTRACT FROM AUTHOR]

Data from 118 members of three consecutive cohorts of students at a private Southern medical school are employed to develop a discrete parameter stationary Markov chains model of the medical specialty choice process. This model was then tested on independent subsamples from the same population, generating a series of predicted specialty choices which did not differ from observed choices by more .than chance expectations. Suggestions for future research, beyond the present demonstration of the utility of a simple Markovian model of a choice process, are offered. [ABSTRACT FROM AUTHOR]

This paper expands the diffusion of information models developed by Funkhouser and McCombs (1972) to include situations involving simple interaction processes and more complicated situations involving both mass mediated messages and interactively mediated messages. This paper develops discrete time models of information diffusion as opposed to the continuous time models developed by Coleman (1964) and Bartholomew (1967) and others. [ABSTRACT FROM AUTHOR]