Future Work

In future studies, the focus should be more on optimizing the balance between time and model complexity to enhance the efficiency of model development in credit scoring. This can be achieved by:

1. Hyperparameter Tuning: Implement methods such as grid search or Bayesian optimization to fine-tune the hyperparameters of the base classifiers (SVC and XGBoost) and the final classifier (Logistic Regression). This process will help to maximize the performance of the model by identifying the optimal settings for each component.

2. Domain Expansion: Apply the developed stacking model to other financial prediction problems, such as loan default prediction and customer churn prediction. This will help to evaluate the model's generality and adaptability across different financial contexts, ensuring its robustness and versatility.