Credit portfolio stress tests that distinguish marginal risk from clustered defaults.
Open the online report · 中文说明 · Model and tail math · Data contract · Case study · Interview notes
Scenario assumptions are explicit. All scenarios reuse the same global, sector and obligor-level random drivers. Multiple loans to one borrower share one default event. The resulting report includes analytic expected loss, simulated mean and sampling error, discrete VaR/ES, paired scenario differences and additive sector attribution.
git clone https://github.com/dev-belly/StressAtlas.git
cd StressAtlas
python -m pip install -r requirements-replay.txt
python -m pip install -e .
stressatlas demo --out outputs
stressatlas verify --out outputs
python -m unittest discover -s tests -vOpen outputs/report.html in a browser. It works offline and lets you switch between loss histograms and sector contributions. Python 3.11+; NumPy and SciPy. Locked numerical versions are supplied for saved-artifact replay.
stressatlas demo --paths 100000 --seed 42 --out outputs-large
stressatlas run --inputs demo/inputs.json --out outputs160 loans · 80 obligors · 4 sectors · CNY 169.6m base EAD · 20,000 shared paths. All PD/LGD/EAD inputs are synthetic assumptions, not calibrated borrower estimates.
| Scenario | Analytic EL / CNY m | 99% VaR / CNY m | 99% ES / CNY m |
|---|---|---|---|
| Baseline | 3.366 | 15.140 | 18.476 |
| Recession | 8.692 | 29.289 | 33.602 |
| Severe | 19.270 | 51.326 | 57.176 |
| Independent defaults | 3.366 | 8.718 | 9.621 |
| Higher asset correlation | 3.366 | 22.085 | 27.706 |
The last two scenarios keep marginal PD, LGD and EAD fixed. Analytic expected loss is unchanged; this simulated portfolio's tail changes substantially. Correlation changes are not guaranteed to increase every individual path or every portfolio quantile.
Source: saved scenario table, sector attribution, normalized inputs, summary and driver fingerprint.
The separate precision report resamples complete paths with the same indices across all scenarios. It recomputes mean loss, VaR, exact empirical ES and their paired changes, reporting approximate percentile intervals and bootstrap SE. These diagnose sampling variation under fixed model assumptions; sparse and discrete tails can make coverage unreliable.
stressatlas precision --inputs demo/inputs.json --resamples 300 --out outputs/precision
stressatlas verify-precision --out outputs/precisionOnline precision report · Method and worked example · Interval rows · All resamples
- Borrower identity: splitting a loan does not create independent default events.
- Common random numbers: compare scenario losses path by path; report the standard error of paired differences.
- Discrete tails: ES integrates exactly the worst
N × (1−α)path-equivalents, including fractional boundary mass. Averaging every loss>= VaRcan be wrong when losses are tied. - Attribution: sectors share the portfolio-tail weights; boundary ties share weight equally. Contributions sum to portfolio ES and do not depend on tie order.
- Numerical checks: exact PD=0/1, analytic EL agreement, scenario monotonicity under fixed correlation, seed/order/chunk invariance and rehashed-report corruption regressions.
inputs.json and the saved seed regenerate all paths. loss_sample.csv contains only the first 200 paths; it is not the full tail sample. The manifest checks six artifacts; verification also reruns the model. Its hashes do not authenticate external source data.
The model has one-period Gaussian factors with deterministic scenario LGD/EAD. It does not model rating migration, contagion networks, dynamic recovery, cash flows or macro calibration. It is not an implementation of Basel regulatory capital. Mean Monte Carlo SE is not a VaR/ES confidence interval.
The random driver bank is held in memory (O(paths × obligors)); chunking bounds the latent-loss intermediates, not the whole simulation's memory. There is no claim of production-scale throughput. MIT license.