LTV Prediction for Financial Institutions
This project develops a predictive tool for assessing loan-to-value (LTV) ratios, a critical financial metric used by banks to evaluate lending risks associated with mortgages.
With high LTV ratios typically indicating higher-risk loans, the project aimed to create a sophisticated model that can accurately compute LTV ratios based on behavioral machine learning techniques, thereby aiding financial institutions in managing credit risk effectively
-
Data Analysis and ML Development: Jupyter Notebook, Google Colab, Visual Studio Code
-
Libraries: Pandas, Numpy, Matplotlib, Seaborn, Scikit-Learn, PyCaret
-
API Development and Deployment: FastAPI, Uvicorn, Codecs