Branch and Bound Method in Feature Selection Process for Models of Financial Condition Evaluation

Abstract

Uncertainty and risk which are associated with company’s activity require a special instrument supporting process of the managers’ decision making. The objective of the research is to determine the number features characterizing the financial risk – financial condition of companies. A quality of financial condition evaluation depends on the selection of variables (features) and criteria of the assessment. The choice of financial ratios in the study of financial standing of companies is crucial. The article presents the proposal to apply branch and bound method to choose sub- optimal subsets of financial ratios that best describe the subject of the research, which is the company, under the assumption that classification algorithms are used for evaluation function creation. The aim of this study is to present a solution that allows the selection of financial ratios with a very high cognitive value, enabling the building of integrated measures assess the financial condition of the company.