Neway and fuzzy methods for modeling diagnostics and predicting the risk of bankruptcy of corpor
Please sign in so that we can notify you about a reply
The monograph is devoted to the general principles of improving neural networks of modeling diagnostics and predicting the likelihood of risk of bankruptcy of corporations in relation to complex modeling conditions: incompleteness, inaccuracies and uncertainty in the data. For the first time, the issues of Bayesovsky regulator of neural networks in the absence of a priori information about the type of noise distribution in the data, assessing the adequacy of neural networks based on the sequential principle of the Wald, the optimal choice of the indicators for the specification of the neural network, reduce the dimensionality of factor space, the assembly of data, aggregate quantitative quantitative quantitative quantitatives and qualitative indicators for their introduction into a neural network model using fuzzy methods. On the basis of the systemic methods of cybernetics, the conceptual basis of neural network modeling, on the basis of which 6 original methods for constructing neural network models of bankruptcy and monitoring the financial condition of corporations in the tasks of ensuring their economic security are proposed. The central method of them is a neural network logistic dynamic method with continuous time, which allows to obtain a forecast for any point in time within the action of a loan agreement.
Theoretical provisions are illustrated in detail on a number of applied tasks. The material of the book is 90% original and summarizes the experience of long -term research of authors on this issue.
The monograph is intended for students, undergraduates and teachers of a wide range of specialties of the information and economic profile of universities, as well as scientists who show interest in the problems of neural network modeling in the field of economics in conditions of high uncertainty
Theoretical provisions are illustrated in detail on a number of applied tasks. The material of the book is 90% original and summarizes the experience of long -term research of authors on this issue.
The monograph is intended for students, undergraduates and teachers of a wide range of specialties of the information and economic profile of universities, as well as scientists who show interest in the problems of neural network modeling in the field of economics in conditions of high uncertainty
Author:
Author:Beloliptsev I.I., Farkhieva S.A., Gorbatkov S.A.
Cover:
Cover:Hard
Category:
- Category:Business & Money
ISBN:
ISBN:978-5-907003-09-5
No reviews found