Fuzzy Sets in Management, Economics and Marketing

The following sections are included:

• CONTENTS

• EDITORIAL

Uncertainty, which is all the time becoming more obvious within social life, economy and management, makes it progressively more difficult to use numerical mathematics for arrivng at a solution to the multiple problems of decision appearing in financial activities. The attempt to incorporate non numerical mathematics in order to be able to establish a new theory of decision in uncertainty [1] has opened up for research in the different areas of knowledge, such as, banking knowledge. The reformulation of concepts such a relation, assignment, grouping and order have been the starting out point for new lines of research which we feel will be most fruitful. In conjunction with Professor Kaufmann [4] we have worked, for many years, in order to construct a body, as homogenious as possible, able to allow optimum groupings of physical or mental objects. For this we drew up what, over the years, has become to be known as the theory of affinities. In this theory, and as a culmination, we introduced the Galois lattices, a simple and elegant means of presenting, in a wholly structured manner, those groups which had been previously found. On this occasion what we intend to do is to present certain algorithms that are capable of generalising the latticed schemes by means of configurations that do not require the conditions of a lattice. In this way we feel that financial decisions will be made easier by the presence of structured and ordered information in a manner that is adequate for the requirements of the decision subject.

Recently we have developed a method to solve a linear goal-programming problem whose parameters are crisp and where both the constraints and the achievement of goals are flexible. In this paper we use this method to solve a model for evaluating a hospital service performance. The Decision Maker usually offers information that allows us to establish suitable values, in fuzzy terms, of the degree to which the goals and constraints are reached. Without distinction between goals and constraints, we will accept that a solution may verify each one with a certain degree of achievement, and the membership function of each fuzzy set describing them represents this degree. Then, we propose as problem solution the decision vectors x that maximize a global measure for the degree of achievement of goals and constraints. Using linear membership functions and, as aggregate function, the weighted mean, the initial fuzzy problem was converted into an equivalent goal programming problem that we solve in an interactive way. Our proposal is applied here to present a model to design the real performance of a surgical service at a local general hospital.

In this paper we introduce a Borda-type decision procedure taking into account agents' intensities of preferences by means of linguistic labels. The advantages of the classic Borda count hold, and are even improved, because the flexibility of gradation increases.

Within the field of multicriteria decision aid (MCDA), sorting refers to the assignment of a set of alternatives into predefined homogenous groups defined in an ordinal way. The real–world applications of this type of problem extend to a wide range of decision–making fields. Preference disaggregation analysis provides the framework for developing sorting models through the analysis of the global judgment of the decision–maker using mathematical programming techniques. However, the automatic elicitation of preferential information through the preference disaggregation analysis raises several issues regarding the impact of the parameters involved in the model development process on the performance and the stability of the developed models. The objective of this paper is to shed light on this issue. For this purpose the UTADIS preference disaggregation sorting method (UTilités Additives DIScriminantes) is considered. The conducted analysis is based on an extensive Monte Carlo simulation and useful findings are obtained on the aforementioned issues.

The private life insurance industry in Greece, as well as in the European region, is rapidly changing since it is challenged with new immigration, demographic and occupational circumstances. This paper is an attempt to demonstrate the use of Hierarchical Classification, a Statistical Multidimensional Data Analysis clustering method, implemented in a graphical environment using software built by the authors. The analysis of 1234 contracts based on 145 variables lead us into a collection of distinct, homogenous contract groups. We believe the method allows to efficiently obtain information about the portfolio picture of the life insurance company with no a piori assumptions and limitations.

The extrinsic quality cues of certification and packaging (bottling) and the intrinsic quality cues of aroma and taste are assumed to influence consumer preferences and the decision to purchase the product. Prior research suggests that, for most utilitarian products, intrinsic cues are more important than extrinsic cues. However, evidence suggests that this hypothesis might not be valid for image products. Quality wine, which is an image product, to a great extent, is used for examining consumer preferences towards these two sets of cues. A sample of 744 Greek wine consumers is used to assess the factors influencing consumer attitudes towards these quality cues. An ordered probit model with sample selectivity reveals that these quality cues are valued by consumers and are perceived to be highly important. Furthermore, they possess different socioeconomic and demographic characteristics. Results support the hypothesis that the market is highly fragmented where the importance of extrinsic and intrinsic quality cues on purchasing behaviour is concerned. The type and source of information received by consumers, their place of origin, disposable income, education and marital status all exert an independent effect on attitude formation. The use of quality cues such as certification, bottling, aroma and taste may be potentially useful in creating niche markets and advancing rural localities.

Consumers' preferences, attitudes and perceptions are examined with respect to fruit juices in Greece. An overview of the European and Greek Juice markets are presented. In an attempt to identify the main criteria and determining consumer purchasing behavior. An extensive number of consumers (800) was interviewed. Data analysis and multicriteria methodology is employed in order to retrieve and determine new market trends.

The principle objective of this research was to provide an analysis and an assessment of both the market for Cretan agricultural products and of the latter's distribution in the local tourist market. A secondary objective was to provide information about Cretan products to foreign clients and to also guarantee the variety and quality of these commodities thereof. The identification of potential new markets and the promotion of Cretan products, adapted to consumer attitudes and beliefs, was to be a further objective of this research. Correspondence Analysis and Non-linear Principal Component Analysis were implemented. The former was conducted to identify the existing relationships among variables, and the latter to provide a thorough examination of the following factors: an analysis of the distribution channels, consumer behaviour, product characteristics and consumer willingness to buy the aforementioned products.

The programming models for selecting fixed income portfolios that are usually proposed in the literature have a number of limitations. For example, they consider profit and immunising risk to be mutually exclusive or suppose a clearly defined planning horizon. In this paper we propose the use of fuzzy programming to solve these limitations and to formalise the problem in a more realistic and flexible way.

This paper examines the relationship between the interest rate, exchange rate and stock price in the Jakarta stock exchange. This was felt timely, as the Indonesian economy is under-going difficult times and there are numerous and conflicting reports on the effect of interest rate and exchange rate on the stock market price. The study was conducted for a five year period from 1993 to 1997 which was divided into three sub periods. Depending on the sub periods being considered, sporadic unidirectional causality from closing stock prices to interest rates and vice versa and weak unidirectional causality from exchange rate to stock price were found. The overall evidence, however, failed to establish any consistent causality relationships between any of the economic variables under study. Hence it seems that Jakarta market efficiently incorporated much of the interest rate and exchange rate information in its price changes at closing stock market index. These results can be used as a measure of stock market efficiency, however with caution, as there are many other dimensions that have to be studied before arriving at any definite conclusion about the efficiency.

Stock price forecasting constitutes a challenging research area. Diverse schemes (such as regression models, neural networks, neurofuzzy systems etc.) have been developed and applied; yet, the overall prediction behavior of such systems is questionable in real world conditions. The major reason limiting the accurate stock price predictions is the existence of a plethora of interrelated agents (quantitative and qualitative) affecting stock price movements and fluctuations. Fuzzy Cognitive Maps (FCMs) seem to constitute a useful modeling tool for the development of a forecasting model, which takes into account the characteristics of the stock market. The main purpose of this work is thus to present analytically the FCM operation mode and the potential extensions of the underlying inference mechanism, and to describe possible applications of FCMs in the domain of stock market.

We compare the out-of-sample performance of monthly returns forecasts for two indices, namely the FAZ and the FT. A linear and a nonlinear artificial neural network (ANN) model are used to generate out-of-sample competing forecasts for monthly returns. We consider two fundamental variables as the explanatory variables in the linear model and the input variables in the ANN model, namely the trading volume and the dividend. The comparison of out-of-sample forecasts is done on the basis of forecast encompassing. The results suggest that the out-of-sample ANN forecasts encompass linear forecasts of both indices. This finding indicates that the inclusion of nonlinear terms in the relation between stock returns and fundamentals is important in out-of-sample forecasting.

In this paper is explored the suitability of genetic algorithms for constructing fuzzy rule bases, as part of an hybrid decision support architecture, involving neural networks for wavelet-filtered daily stock rates of change. Specifically, the main structure of the suggested methodology combines a wavelet-based noise removal system, a "multilayer perceptron feedforward neural network" and finally a fuzzy system, which provides the trader with both, linguistic and numerical output, representing a buy/hold/sell strategy. The use of wavelet filtering in data pre-processing, improves the predictability of neural networks, however, it involves the selection of proper wavelet bases. Therefore, by applying genetic algorithms in fuzzy rule bases for optimizing the decision policy, the paper aims at offering a decision support, independent of the selection of the wavelet basis. It is also demonstrated how, based on the test results, the overall system is able to make successful trend prediction, which is then used to create an output similar to the policy that traders would apply if forward price movement was considered to be known.

In this article we propose a model for measuring the adequacy of business decisions with a given standard or set of standards they are designed to achieve. These standards or set of standards can either be external contraints or self-imposed objectives (internal constraints). The model, which is presented in the first part, is based on the theory of expertons. The model is then tested on a sample of 161 companies. Finally, the results are presented and discussed.

In this article we present a new fuzzy methodology to determine multiple IRRs, based on J.T.C. Mao's algorithm. Also, we present an alternative algorithm that exhibits a high level of efficiency and efficacy to solve the multiple IRR problem. The analysis and algorithms presented here have not been reported so far in the fuzzy literature.

The aim of this article is to propose an example-based intelligent approach for classifying enterprises into different categories of credit risk. The data used are of both numerical and linguistic nature. The methodology used for the rule based categorization task is the well-known inductive machine learning approach, based on entropy information. The drive for this paper was the application domain, which is very common in banking management worldwide, moreover very often turns to be a confusing and time-consuming situation. The goal is to obtain a model that correctly classifies a training sample of 130 enterprises with 76 decision variables to the predetermined classes, using a substantially less amount of attributes, trying at the same time to minimize the error rate. Data are transformed into a proper input database, and then training experiments take place in order to find the optimal settings for the training phase. After having obtained the right adjustments, the training phase initiates. The aim is to produce a comprehensible decision tree as an output, which can then be transformed to a set of simple IF/THEN rules. The specific decision tree produced from the training data, uses only 16 attributes to be formed, and is equivalent to a few comprehensible and short rules consisting of 2 to 10 premise parts. As a result, the classification task is made easier to perform and check, the amount of required data is minimized, and finally the whole process is easier to use in decision-making. Furthermore, the produced decision tree works as a knowledge generator and thus, reflects the banking organization's expertise on the application domain, represented by a handy and meaningful set of rules. Finally, the classifier could be continuously reformed, by adding to it every new credit-risk case, becoming a more and more accurate and robust classification model with time.

The uncertainty that prevails in the financial and investment environment has prompted banks and other financial institutions to seek out greater efficiency in the management of their assets and liabilities. Today's asset management decisions create tomorrow's problems as well as tomorrow's opportunities. This need has led to studies concerning the optimal balance among profitability, risk, liquidity and other uncertainties. The present paper makes a brief overview of the bank ALM techniques that have been developed and used over the last 20 years.

The following sections are included:

• Author Index