Selecting the right attributes for a credit scoring model: The chi-square test

Link to Credit scoring in microfinance and banking:
https://www.youtube.com/watch?v=ZDVMOsoXsz4

Often, we base our credit scoring models on statistical analysis of historical data. We try to find a statistical relationship between an attribute value and a default. An attribute is a characteristic of the loan or the client. Examples of attributes are gender, business activity, age and so on. We could find that certain business activities or age groups tend to default more often. This is useful information that we can include in our scoring model. Both scoring approaches use historical loan data. It could also be that there is no relation between the attribute and the defaults. Of course, we are looking for those attributes that have the most predictive power. In this video, we explain how to select the right and the most meaningful attributes for a scoring model. We use the chi-square test for this purpose. We show step by step how to apply this test in practice and provide an Excel example. (You can download the Excel example at: www.stachanov.com/downloads)

tags:
stachanov,André Koch,Excel,Scoring,Chi-square,Chi-squared,testing,statistical relevance,credit scoring,risk management,chi-square distribution,bank financial management,notation,attribute,selecting attributes,Excel example,how-to,Statistical analysis

Видео Selecting the right attributes for a credit scoring model: The chi-square test канала Stachanov Holding B.V.