Estimering av kreditbetyg - Syntetisk rating med Ordered Probit Model

Abstract

Calculations of net present value are based on a discount rate, which requires knowledge about the company’s cost of debt. A common way to do this is by observing the company’s credit rating, when no credit rating is given it’s possible to use a synthetic rating model based on quantitative data to estimate a credit rating. Thus, the aim of this study is to increase the understanding of which quantitative data, in terms of key figures, that can be used to explain the given credit ratings for companies from the Nordic countries. In developing the model, key figures found significant in previous studies are being tested for whether or not they are applicable for companies from the Nordic countries. Also, this study seeks to find if the addition of variables based on trends adds explanation to credit ratings. An Ordered Probit Model is used on data from all Nordic companies, the financial sector excluded, with credit rating from Standard & Poor’s. The model correctly assigns a credit rating to 57 percent of our sample firms, similar to findings in previous studies. Long term debt leverage and operating margin is found significant in estimation of credit ratings. This is in accordance with previous studies on other markets, and the only known study on Swedish companies. Furthermore, a company’s total assets is shown to be significant, something that’s only partly displayed in the previous study on Swedish companies. Finally, it’s presented that growth of total income may contribute to the explanation of credit ratings.