Credit Portfolio Management of Corporate and Commercial Loans with Robust Machine Learning Methods
Bibliographic record
Abstract
This thesis addresses critical challenges in credit portfolio risk management for financial institutions, focusing on corporate and commercial loan portfolios. This thesis comprises three main project. The first two projects are published in the Journal of Credit Risk the third project has been accepted for publication at the Journal of Applied and Numerical Optimization: 1. Credit Risk Rating Modeling: This study introduces a novel approach to credit risk rating for corporate entities, enabling banks to make accurate, cost-effective rating decisions while maximizing risk-adjusted returns under model uncertainty within the Basel Regulatory Framework. By leveraging ordinal information from expert-assigned credit ratings, the methodminimizes expected economic costs. Empirical results demonstrate higher returns on regulatory capital for medium-sized North American companies. 2. Distributionally Robust Optimization (DRO) in Credit Risk Management: This project applies a DRO framework to enhance credit risk management by addressing data uncertainty and model misspecification. Two applications are explored: predicting significant increases in credit risk (SICR) under the IFRS 9 Expected Credit Loss framework and managing risk limits for corporate loans. The findings show that DRO improves model robustness by accounting for distributional uncertainty, supporting more informed and regulatory-compliant decision-making. 3. Robust Contextual Bandit Method for Optimal Loan Offering: This study proposes a group-DRO-enhanced, doubly-robust contextual bandit approach to optimize loan product offerings. Tailored for high-stakes lending decisions, this method leverages historical data to design policies while mitigating biases and uncertainties. By incorporating group-based ambiguity sets and fairness constraints, such as demographic parity or equal opportunity, the approach ensures robustness against worst-case shifts in sensitive subgroups and aligns with ethical and regulatory standards. Empirical evidence from a small business credit card portfolio demonstrates significant improvements over standard methods, advancing responsibleAI in finance. Collectively, these contributions provide advanced methodologies to enhance credit risk management, improving the modeling and management of corporate and commercial loan portfolios for financial institutions.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.002 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.005 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".