About the Project


What is Credit Risk Modeling?



Put simply, credit risk modeling relates to the task of predicting whether a loan will default based on the variables related to the loan itself. Such a task is rather important in the finance industry, due to the capital at risk in the loan. Over the last few decades, the utilization of data science and algorithms to predict a loan recipient's outcome have become immensely relevent. Data science approaches allow banks to utilize unbiased methods to approve, and not approve, of potential loans based on what the models probabilistically display. If you're interested in knowing more about credit risk modeling, here is a more extensive discussion on what credit risk modeling really is.


What's under the hood?



What you interact with here is a pre-trained model placed into production utilizing Heroku and the python web-framework, Django. Through the utilization of these tools, I have been able to deploy a random forest classification model based on data provided by Dr. Brent Albrecht. This model was iteratively trained through an extensive process which resulted in the variables you see on the model page. This model works to utilze your input, make a probability prediction of default, and returns it to you. If you'd like to read my full report you can find it here.
It is worth mentioning this dataset being based in the past and most likely does not reflect current loan environments.