The Impact of Pseudo-Science in Financial Loans Risk Prediction
This addresses fairness and bias issues in financial lending for society, highlighting incremental insights into survival bias dynamics.
The study tackled the problem of pseudo-scientific assumptions and survival bias in machine learning models for financial loan risk prediction, finding that socially optimal models may not significantly reduce accuracy, while models with survival bias create an illusion of improvement as accuracy slightly deteriorates but recall and precision improve over time.
We study the societal impact of pseudo-scientific assumptions for predicting the behavior of people in a straightforward application of machine learning to risk prediction in financial lending. This use case also exemplifies the impact of survival bias in loan return prediction. We analyze the models in terms of their accuracy and social cost, showing that the socially optimal model may not imply a significant accuracy loss for this downstream task. Our results are verified for commonly used learning methods and datasets. Our findings also show that there is a natural dynamic when training models that suffer survival bias where accuracy slightly deteriorates, and whose recall and precision improves with time. These results act as an illusion, leading the observer to believe that the system is getting better, when in fact the model is suffering from increasingly more unfairness and survival bias.