Logical-linguistic method for assessing the risk of specialized lending (on the example of project financing)

Abstract

Credit risk assessment is associated with uncertainty of many factors which are hard to describe mathematically. This reduces the probability of success of specialized lending as well. Among the various risk assessment methods suitable for specialized lending risk assessment, it’s possible to single out a scenario approach that uses the methods of fuzzy set theory to calculate the values of membership functions. However, the problem of ranking the set of project implementation scenarios for which a loan is requested is not studied enough. Despite the existing technical difficulties of risk factors management, the ranking of the risks of project implementation can be carried out by comparing with the ones from the database using the logical-linguistic classification method. The purpose of the study is to develop new methods of credit risk assessment based on the logical and linguistic classification of specialized lending projects as well as algorithms for assessing the credit risk of the project with determining its rating. The definition of a Project is used according to the concept of Project Finance as one of the classes of specialized lending. Methods: the assignment of weight factors for all subclasses of credit requirements, the assignment of risk levels for all values of the types of criteria, introduction of a table of the risk value assessment or the creditworthiness of the project, the formation of a database of reference attribute strings for all classes of projects, generating attribute strings for the analyzed project, assigning the analyzed project to the benchmark using the logical-linguistic classification method, assigning risk degree values for the analyzed project for all values of the criteria (attribute) types of the selected reference project, calculating the credit risk assessment of the analyzed project and determining the rating of the analyzed project for decision making. Results: based on the proposed risk assessment method, an algorithm for calculating the risk assessment of lending to the analyzed project with determining its rating was developed. Practical significance: the results of the research can be used in the development of computer program allowing to accelerate risk assessment of project financing.