<p>The new accounting standards IFRS and IAS was implemented in January 2005, the purpose was to harmonize accounting standards world wide. As a result of these new rules the real estate companies are allowed to appraise real estates assets at market value. It has long been debated whether this appraisal fulfils the requirements for relevance, reliability and comparability.</p><p>The paper attends to how banks control the information given by the real estate companies in the process of credit granting. As an introduction we describe the purpose of accounting and the problematic´s about appraisals with market values. Further on there is a discussion on which factors are taken into consideration when banks grant credits.</p><p>The study is built on interviews at three different commercial banks. At each bank one bank advisor was interviewed, the respondent were given specific questions and at the same time were free to speak openly about the subject.</p><p>In conclusion all banks focus on the real estates companies ability to repay the bank credits with revenues from the investment. With the focus on reimbursement banks will minimize the risk of credit losses and thereby decrease the risk of history repeating itself with new bank crisis. Besides the ability to repay credits banks also consider the fact’s, who is leading the company and how this owner has conduct prior businesses. In comparison whit theory and empirics all banks, in there credit granting process, fulfil the requirements for relevance, reliability and comparability. In summary the banks do not consider real estates valued at market prices as a problem in there assessments.</p>

The new accounting standards IFRS and IAS was implemented in January 2005, the purpose was to harmonize accounting standards world wide. As a result of these new rules the real estate companies are allowed to appraise real estates assets at market value. It has long been debated whether this appraisal fulfils the requirements for relevance, reliability and comparability. The paper attends to how banks control the information given by the real estate companies in the process of credit granting. As an introduction we describe the purpose of accounting and the problematic´s about appraisals with market values. Further on there is a discussion on which factors are taken into consideration when banks grant credits. The study is built on interviews at three different commercial banks. At each bank one bank advisor was interviewed, the respondent were given specific questions and at the same time were free to speak openly about the subject. In conclusion all banks focus on the real estates companies ability to repay the bank credits with revenues from the investment. With the focus on reimbursement banks will minimize the risk of credit losses and thereby decrease the risk of history repeating itself with new bank crisis. Besides the ability to repay credits banks also consider the fact’s, who is leading the company and how this owner has conduct prior businesses. In comparison whit theory and empirics all banks, in there credit granting process, fulfil the requirements for relevance, reliability and comparability. In summary the banks do not consider real estates valued at market prices as a problem in there assessments.

This thesis focuses on application of artificial intelligence techniques in credit risk management. Moreover, these modern tools are compared with the current industry standard - Logistic Regression. We introduce the theory underlying Neural Networks, Support Vector Machines, Random Forests and Logistic Regression. In addition, we present methodology for statistical and business evaluation and comparison of the aforementioned models. We find that models based on Neural Networks approach (specifically Multi-Layer Perceptron and Radial Basis Function Network) are outperforming the Logistic Regression in the standard statistical metrics and in the business metrics as well. The performance of the Random Forest and Support Vector Machines is not satisfactory and these models do not prove to be superior to Logistic Regression in our application.