Mitigating Default Risk in the Consumer Credit Market

This paper aims to evaluate recent policy updates in a credit scoring model and determine if the new model is efficient, as well as further investigate other potential risk factors. In order to evaluate the policy changes, the proprietary dataset is first categorized and estimated by a logistical regression model and secondly the dataset is transformed according to new policies and then simulated in a second regression. The choice of variables is further tested to ensure robust result of the identified risk factors and best fitting of the model. The discoveries points towards efficient implemented policy changes to the scoring model, and the identification of other potential risk factors which leads to a set of managerial suggestions.