| Title |
Comparison of relevant credit risk assessment algorithms |
| Translation of Title |
Aktualių kredito rizikos įvertinimo algoritmų palyginimas. |
| Authors |
Rimašauskas, Simas ; Belovas, Igoris ; Gricius, Rolandas |
| DOI |
10.15388/LMR.2025.44496 |
| Full Text |
|
| Is Part of |
Lietuvos matematikos rinkinys. Ser. A.. Vilnius : Vilniaus universiteto leidykla. 2025, t. 66, p. 39-51.. ISSN 0132-2818. eISSN 2335-898X |
| Keywords [eng] |
credit risk, machine learning ; XGBoost ; LightGBM ; AdaBoost ; logistic regression ; random forest |
| Abstract [eng] |
Assessing credit risk is essential when making financial decisions, especially investing in debt securities. As the bond market is the largest securities market in the world, a great demand exists for tools to assess issuer creditworthiness. Furthermore, information describing the probability of default is also useful in other areas, such as risk management. In the era of machine learning and big data, new techniques have emerged that allow for automated risk assessment based on large amounts of data. Traditional creditworthiness assessment methods may be inaccurate, as the investor could be biased or misinterpret available information. A review and comparison of modern tools that would allow intelligent processing of large amounts of information will help assess the issuer's credit risk as objectively as possible. |
| Published |
Vilnius : Vilniaus universiteto leidykla |
| Type |
Journal article |
| Language |
English |
| Publication date |
2025 |
| CC license |
|
