Laurent Donzé

Fuzzy methods for customer relationship management and marketing : applications and classifications

Information overload has made it increasingly difficult to analyze large amounts of data and generate appropriate management decisions. Furthermore, data is often imprecise and will include both quantitative and qualitative elements. For these reasons, it is important to extend traditional decision making processes by adding intuitive reasoning, human subjectivity, and imprecision. Fuzzy Methods for Customer Relationship Management and Marketing: Applications and Classifications explores the possibilities and advantages created by fuzzy methods through the presentation of thorough research and case studies. This book covers a variety of possible fuzzy logic approaches to customer relationship management and marketing, making it a valuable resource for not only students and researchers but also executives, managers, marketing experts, and project leaders who are interested in applying fuzzy classification to managerial decisions.

Impact of Swiss technology policy on firm innovation performance: an evaluation based on a matching approach

This paper investigates the impact of the promotional activities of the Swiss Commission of Technology and Innovation (CTI) on the innovation performance of the supported firms based on a matched-pairs analysis of 199 firms supported by the CTI in the period 2000–2002. CTIs promotional activities significantly improved the innovation performance of the firms that they supported with respect to six different measures of innovation performance. This could be shown by four different matching methods. A further finding was that the magnitude of the impact correlated positively with the relative size of the financial support, as measured by the quotient of the volume of financial support to the volume of a supported firms own research and development expenditures. Copyright , Beech Tree Publishing.

Statistical Characteristics of Distributions Obtained Using the Signed Distance Defuzzification Method Compared to Other Methods

Having in mind the evaluation of linguistic questionnaires, we aim to present a comparison in terms of statistical measures between on one side a relative recent defuzzification method, known as the signed distance method, and on the other side, other well-known traditional methods. The distribution’s properties of data resulting from the defuzzification process are generally not given or investigated. By simulations, we intend to investigate the location, dispersion and symmetry characteristics of the estimated distributions. Our simulations for different cases of input distributions and membership functions show first that the computed statistical measures don’t depend on the sample sizes. This phenomenon is particularly remarkable in the case of the signed distance and the mean of the maximum methods. Second, the signed distance is the method tending the most to conserve the symmetry of the distributions while the smallest and largest of maximum are the worst in keeping the skewness property.

Individual and Global Assessments with Signed Distance Defuzzification, and Characteristics of the Output Distributions based on an Empirical Analysis

Considering a particular kind of questionnaires called linguistic questionnaires, we apply fuzzy logic to provide individual and global weighted evaluations in the case of the signed distance defuzzification method. We test our method on real data coming from a survey of the financial place of Zurich (Switzerland). Furthermore, we have been enable to give a look at the output distributions, and put into evidence their statistical properties. In particular, normality of distributions draws our attention. One of our main findings is that the individual evaluations calculated with the signed distance defuzzification method tend to be normally distributed.

Central Moments of a Fuzzy Random Variable Using the Signed Distance: A Look Towards the Variance

The central moments of a random variable are extensively used to understand the characteristics of distributions in classical statistics. It is well known that the second central moment of a given random variable is simply its variance. When fuzziness in data occurs, the situation becomes much more complicated. The central moments of a fuzzy random variable are often very difficult to be calculated because of the analytical complexity associated with the product of two fuzzy numbers. An estimation is needed. Our research showed that the so-called signed distance is a great tool for this task. The main contribution of this paper is to present the central moments of a fuzzy random variable using this distance. Furthermore, since we are interested in the statistical measures of the distribution, particularly the variance, we put an attention on its estimation using the signed distance. Using this distance in approximating the square of a fuzzy difference, we can get an unbiased estimator of the variance. Finally, we prove that in some conditions our methodology related to the signed distance returns an exact crisp variance.

Testing Fuzzy Hypotheses with Fuzzy Data and Defuzzification of the Fuzzy p-value by the Signed Distance Method.

We extend the classical approach of hypothesis testing to the fuzzy environment. We propose a method based on fuzziness of data and on fuzziness of hypotheses at the same time. The fuzzy p-value with its α-cuts is provided and we show how to defuzzify it by the signed distance method. We illustrate our method by numerical applications where we treat a one and a two sided test. For the one-sided test, applying our method to the same data and performing tests on the same significance level, we compare the defuzzified p-values between different cases of null and alternative hypotheses.

Impact of the density support on the matching bias: a matched-pair analysis based on business survey data

Matching methods have been extensively used to evaluate economic policy. However, they are not without fault and, indeed, a selection bias may appear. In 1998, Heckman and al. (1998) have precisely characterize this bias by a decomposition in three parts which can be non parametrically estimated. These estimations depend among others on the support of the conditional density function of the covariates X used for the matching. In a study of the impact of the policy of supporting the adoption of advanced manufacturing technologies by Swiss firms, we have measured the different components of the bias. We put into evidence the extreme sensibility of the results to the measure of support density.

Aufbau, Weiterentwicklung und Pflege des KOF/ETH-Unternehmenspanels

Im Rahmen des SPP „Zukunft der Schweiz“ (Modul Arbeitswelt) sollte ein Befragungsinstrument geschaffen werden, das eine Dauerbeobachtung des soziookonomischen Wandels im Unternehmenssektors, insbesondere hinsichtlich struktureller Veranderungen der Arbeitsnachfrage, ermoglicht. Zu diesem Zweck baute die KOF/ETHZ ein Unternehmenspanel auf, das reprasentative Aussagen fur den gesamten privaten Wirtschaftssektor erlaubt. Das Panel beruht auf einer nach Branchen und Grossenklassen geschichteten Stichprobe von anfanglich uber 7000 Firmen. Die entsprechende Adressdatei, die neben zahlreichen Strukturmerkmalen der Firmen alle fur einen reibungslosen Ablauf von Panelbefragungen (spezifische Anlaufstellen, Files zur Kontrolle der Antworten, Mahnaktion und Nichtbeantworterbefragungen, usw.) erforderlichen Daten enthalt, wurde im Laufe des Projekts regelmassig weiterentwickelt und aufdatiert. Basierend auf dieser Adressdatei wurden 1998 z.Hd. von zwei anderen Projekten des Verbunds zwei Panelbefragungen („Qualifikation“ Anfang 1998, „Internationalisierung“ Herbst 1998) durchgefuhrt. Wahrend die Rucklaufquote der ersten Erhebung nicht zu befriedigen vermochte, erzielten wir bei der zweiten ein sehr gutes Ergebnis, was unter anderem auf die in der Zwischenzeit optimierten Arbeitsablaufe und die Reduktion des Fragebogenumfangs zuruckzufuhren war. Ein Schwerpunkt der Projektarbeiten bestand aus methodischen Untersuchungen insbesondere zur „non response“-Problematik („unit“ und „item“-non- response). Ausgehend von einer Evaluation verschiedener zur Losung dieser Probleme vorgeschlagenen Ansatze wurden spezifische Verfahren ausgewahlt, EDV-massig implementiert (eigene Software-Entwicklung) und im Rahmen des zweiten von der KOF bearbeiteten SPP-Projekts eingesetzt. Es zeigte sich, dass der Einsatz solcher fortgeschrittener statistischer Techniken, mit deren Anwendung wir in der Schweiz z.T. Neuland betraten, die Qualitat deskriptiver und explikativer Analysen wesentlich verbessern kann. Die fur die Anwendung dieser Verfahren erforderlichen Software- Elemente stehen interessierten Kreisen als SAS-Makros zu Verfugung. Der im ursprunglichen Projektplan vorgesehene Aufbau einer Zeitreihendatenbank der Umfrageresultate auf Firmendaten – diese stehen z.Zt. als Serie von Querschnitts-Datensatzen zur Verfugung – musste auf Phase II des Panelprojekts verschoben werden; im Rahmen der gegebenen Ressourcen schien es uns sinnvoll, die methodischen Untersuchungen gegenuber den ursprunglichen Absichten zu forcieren. Insgesamt darf man u.E. festhalten, dass nun ein leistungsfahiges Panel zur Verfugung steht, das in Phase II des SPP in verschiedener Hinsicht weiterentwickelt werden kann (Anpassung der Stichprobe an die Betriebszahlung 1998, Zeitreihendatenbank, Durchfuhrung weiterer Panelerhebungen (auf Kosten der KOF), methodische Untersuchungen im Hinblick auf Panelschatzungen bei Vorliegen von „non response“.