Collateral liquidity level and credit scoring model in retail business loans segment - case of 1 bank in Vietnam | |
| Author | Nguyen Le Hieu |
| Call Number | AIT Project no.PMDS-22-06 |
| Subject(s) | Credit scoring systems Machine learning Credit ratings--Vietnam Consumer credit--Vietnam |
| Note | A project study submitted in partial fulfillment of the requirements for the degree of Professional Master in Data Science and Artificial Intelligence Applications, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Project ; no. PMDS-22-06 |
| Abstract | Credit scoring or probability of default prediction is an essential task in banking institution. There are many papers research on credit scoring model and found that explanatory variables which play role as loan risk predictors are different from country to country, case by case, region by region. Developing countries is still facing with critical information asymmetry which is an obstacle in features selection and therefore affect the quality of loan risk prediction model due to low quality of data and lack of data. This study investigated the effectiveness of adding collateral liquidity level factor to list of prediction factors of probability of default (PD) prediction model which serve bad debt classification task for case of a retail bank in Vietnam, a developing country. In this paper, I examined whether collateral liquidity level has significantly impact on PD of loans, construct PD prediction model with collateral liquidity factor as one of the predictors. Then performance of selected model is compared with the ones proposed by 2 other studies in developing countries. Our key findings are (1) Collateral liquidity level has discriminatory power to classify bad debt and is found to have significant effect on PD of loans, (2) Performance of final model in this model overperform the one in Gool at Al (2011) but worse than Dinh and Kleimeier (2011). Our work can help banking institutions to have more addition of prediction factors choices for loan risk prediction model in retail secured loans for developing countries in order to deal with information asymmetry problem. |
| Year | 2022 |
| Corresponding Series Added Entry | Asian Institute of Technology. Project ; no. PMDS-22-06 |
| Type | Project |
| School | School of Engineering and Technology |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Professional Master in Data Science and Artificial Intelligence Applications (PMDS) |
| Chairperson(s) | Chaklam Silpasuwanchai; |
| Examination Committee(s) | Vatcharaporn Esichaikul;Mongkol Ekpanyapong; |
| Degree | Professional Master in Data Science and Artificial Intelligence Applications - Asian Institute of Technology, 2022 |