A machine learning approach in stress testing US bank holding companies
Bibliographic record
Abstract
This paper assesses the utility of machine learning (ML) techniques combined with comprehensive macroeconomic and microeconomic datasets in enhancing risk analysis during stress tests. The analysis unfolds in two stages. I initially benchmark ML’s efficacy in forecasting two pivotal banking variables, net charge-off (NCO) and pre-provision net revenue (PPNR), against traditional linear models. Results underscore the superiority of Random Forest and Adaptive Lasso models in this context. Subsequently, I use these models to project PPNR and NCO for selected bank holding companies under adverse stress scenarios. This exercise feeds into the Tier 1 common equity capital (T1CR) densities simulation. T1CR is the equity capital ratio corrected by some regulatory adjustments to risk-weighted assets. Crucially, findings reveal a pronounced left skew in the T1CR distribution for globally systemically important banks vis-à-vis linear models. By mirroring distress akin to the Great Recession, ML models elucidate intricate macro-financial linkages and enhance risk assessment in downturns. • ML modeling improves forecast accuracy of PPNR and NCO over linear models. • ML modeling better approximates T1CR density than linear models in turmoil. • Linear models tend to underestimate systemic risk in stress test. • ML modeling provides realistic picture of complex macro financial linkages.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.004 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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".