Multicriteria decision aid classification methods (Q1848995): Difference between revisions
From MaRDI portal
Added link to MaRDI item. |
Set profile property. |
||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank |
Latest revision as of 04:55, 5 March 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Multicriteria decision aid classification methods |
scientific article |
Statements
Multicriteria decision aid classification methods (English)
0 references
28 November 2002
0 references
A decision problem may consist in the choice of a preferred alternative, the ranking of alternatives, the classification or sorting of alternatives into predefined groups or the description of features of the alternatives. The book is concerned with multicriteria decision aid methods for classification/sorting. Classification and sorting are distinguished by the description of groups, which can be either nominal or ordinal. This general outline of the topic and objectives of the book are described in Chapter 1. Chapter 2 provides a review of classification techniques including statistical discriminant analysis, econometric logit and probit analysis as well as non-parametric methods such as neural networks, machine learning, fuzzy sets, and rough sets. In Chapter 3 the concepts of MultiCriteria Decision Aid (MCDA) techniques are introduced. The general MCDA techniques of multiobjective programming, multiattribute utility theory, outranking and preference disaggregation are outlined. For classification problems the authors distinguish methods based on interrogation of the decision maker (analytic hierarchy process, ELECTRE TRI) and preference disaggregation methods. In Chapter 4 the two preference disaggregation techniques UTADIS (utilités additives discriminantes) and MHDIS (multigroup hierarchical discrimination) are discussed in detail. Chapter 5 consists of an experimental study of classification techniques. The performance of seven methods (linear discriminat analysis, quadratic discriminant analysis, logit analysis, rough sets, ELECTRE TRI, MHDIS and two versions of UTADIS) is analysed in a simulation experiment. The analysis shows that the proposed MCDA techniques can be considered as an efficient alternative to classical techniques. Chapter 6 covers real world applications from the finance sector. The problems studied are bankruptcy prediction, corporate credit risk assessment and stock evaluation. Models based on the methods used in the experimental study of Chapter 5 are developed and their results compared. The book is concluded with a summary Chapter 7, quite extensive references and a short index.
0 references
multicriteria decision aid
0 references
classification problem
0 references
MCDA in finance
0 references