Showing papers in "Operations Research and Decisions in 2005"

Effect of dividend and repurchase announcements on the Polish stock market

Abstract: The main aim of this paper is to explore the information content of dividend and buy back announcements. Using daily data from the Warsaw Stock Exchange, we investigate the reaction of stock prices of the announcing firms as well as the industry rivals to the announcement issue. The regression analysis is carried out to explain the differences in magnitude of valuation effects across rivals.

Fuzzy solution of the linear programming problem with interval coefficients in the constraints

Abstract: A fuzzy concept of solving the linear programming problem with interval coefficients is proposed. For each optimism level of the decision maker (where the optimism concerns the certainty that no errors have been committed in the estimation of the interval coefficients and the belief that optimistic realisations of the interval coefficients will occur) another interval solution of the problem will be generated and the decision maker will be able to choose the final solution having a complete view of various possibilities.

Multiple criteria decision support in organizational and management chosen problems solving

Abstract: This article presents an introduction to multicriteria decision making using two decision tools: the Analytic Hierarchy Process (AHP) and its generalization to dependence and feedback – the Analytic Network Process (ANP). The discussion involves theoretical aspects of these methods and some examples of their applications, (e.g. the first application of the ANP, in improving of food quality products, in Poland), in organizational and management problems solving. AHP and ANP introduced by Thomas L. Saaty from the University in Pittsburgh, USA. The Analytic Hierarchy Process has been one of the fastest developing mathematical methods over the recent years used for solving multi-criteria decision problems. The AHP is a general theory of measurement based on some mathematical and psychological principles. In that method a hierarchic decision scheme is constructed, by the breaking the problem into decision elements: goal, criteria, subcriteria, sub-subcriteria (…) and decision alternatives. The goal is on the top of hierarchy, whereas alternatives create the lowest level of hierarchy. The importance of every decision element is established, through the pair-wise comparison of elements on each level of the hierarchic structure, with regard to elements on the level above. To do the comparisons it is using the Saaty’s fundamental scale for paired comparisons for the analysis of both quantitative and qualitative variables. The Analytic Network Process (ANP) is a new theory that extends the Analytic Hierarchy Process (AHP). The basic structures are networks, which allow interactions and feedback within the clusters and between the clusters. So, it can be applied for solving more sophisticated decision problems. Authors’ intention was to showing utility of these methods, which can be successfully applied in the solution of any multicriteria enterprise.

Lower-limit barrier in the problem of teh identification of a barrier in the functioning of a certain inventory storage and issue system

Abstract: The paper investigates a certain inventory system whose input is a non-aggregated dynamicparameter process. The authors derive equations that define the distribution of conditional probabilities for the case of a lower-limit barier in subsystem L. They depend on the parameters of the functioning of transport subsystem and the parameters of the process of product supply to finite-volume storage.

Co-integrating relationship between terms of trade, money and current account : the Italian evidence

Abstract: The paper analyses long-run relationships between terms of trade, money and current account for Italy in the period from the first quarter of 1975 to the first quarter of 2001.

Multivariate analysis of bankruptcy on the example of building industry

Abstract: The main purpose of this research is to examine the different factors affecting the predictive power of bankruptcy models as for example: the way of defining “bankruptcy”, the elimination, or not of the multicollinearity appearing in formalized model, the stability of time-related financial ratios as a stability determinant of formalized and non-formalized models and so on. Some general conclusions were formulated about: effectiveness of formalized and non-formalized model, population homogeneity factor, a model worked out for a multi-industry population and used in single-industry population and so on.

Binomial model of real option valuation

Abstract: Real option valuation methods used in firm valuation process allow taking into consideration firm’s flexibility and its adaptability to environmental changes. In the binomial tree model it is assumed that stock price changes are composed of a great number of small binomial changes. This assumption was first used by Cox, Ross and Rubinstein. In the method, the time period to expiration date is divided into small periods of time ?t. In each period of time ?t share price can change to one of the two values: Su or Sd. Assuming that u > 1, d

Modified Shapley-Shubik power index for parliamentary coalitions

Abstract: Classical power analysis does not involve preferences of players (parties). Classical power indices are constructed under assumption of equal probability of occurrence for each coalition. The paper contains a proposition of relaxation of this assumption, based on extended Shapley–Shubik power index approach.

Real option valuation methods – Black–Scholes models

Abstract: Real option valuation methods used in firm valuation process allow taking into consideration firm’s flexibility and its adaptability to environmental changes The risk is taken into consideration at an expected rate of return in real option – an expected return rate is related to CAPM model Similarly to return rate of stocks, an expected rate of return µ = ? + ? in real option valuation process Dividend rate in real option valuation could be interpreted as the cost of delay in project starting (an income that is refused in exchange of the possibility of rejecting a given option) The Black–Scholes model describes the pricing of European call option for share without dividend payment A standard deviation of yearly return rate of stocks is a risk measure in Black–Scholes model It is assumed that the variability of stocks is constant during the year Actually, the variability is changing with time So, in the paper, it is shown that the variability of stocks at WSE is changing with time and for this reason a short term variability should be taken into account in real option valuation process, ie, the last 90–180 days Other models based on the Black–Scholes one that refuse its basic assumptions are also presented

On different estimators of a population mean based on ranked sets

Abstract: A review of results concerning the problem of sampling based on ranked sets is presented. From an infinite or finite population n independent samples of n elements each are drawn. The samples are ranked and then n elements are chosen to be measured.

Methods of finding the effective portfolio for semi-variance

Abstract: In the classic Markowitz model, risk is measured by the return rates variance. However, equal treatment of negative and positive deviations from the expected return rate is a slight shortcoming of variance as the risk measure. Markowitz defined semi-variance to measure the negative deviations only. However, finding the portfolio with minimum semi-variance is much more difficult than finding a portfolio with minimum variance. The critical line method proposed by Markowitz in 1959 was the oldest method for finding optimum portfolios for semi-variances. That method was highly complicated and as a consequence the search for methods of finding a quasi-optimum solution continued. Quasi-optimum solutions are based on the colower partial moments. Until today they find application in practice. Their advantage is that it is possible to use one of many available software packages for square or non-linear optimization. Unfortunately the solution obtained is quasi optimal and it is not known how far it deviates from the optimum solution. As a consequence, the need to formulate a new method that could offer optimum solution and at the same time would be simple and easy for software design as a means to select optimum portfolios with the minimum semi-variance from the assumer return rate appeared.