Credit Risk Analyst — PD / LGD / EAD modelling · IFRS 9 / AASB 9 · stress testing · model validation
CPA-qualified credit risk analyst with ~4 years across the full credit lifecycle at a private credit lender. I build institutional-grade credit-risk analytics — from PD/LGD/EAD models and IFRS 9 expected credit loss through stress testing, model validation, and portfolio monitoring — on real loan-level and Australian regulatory data.
Based in Sydney · English / Mandarin · Open to credit-risk and quantitative-modelling roles
- Credit risk models: PD, LGD, EAD; IFRS 9 / AASB 9 expected credit loss and staging; credit scorecards (logistic regression + WOE/IV); risk-based pricing
- Stress testing: downturn scenarios, downturn LGD, macro overlays
- Model validation & monitoring: discrimination (AUC, Gini, KS), calibration, population stability (PSI), out-of-time / out-of-regime testing; transition matrices, stage movements, early warning
- Australian regulatory landscape: APRA (APS 112/113/220), Basel PD/LGD/EAD concepts, Pillar 3, and ABS / RBA / APRA public data
- Tools: Python (pandas, scikit-learn), SQL, Power BI, Git
| Project | What it demonstrates |
|---|---|
| mortgage-credit-risk-pd-lgd-ead | PD (logistic, AUC 0.81), real LGD from actual loss data (reconciled to the vendor's own loss field at 0.99), EAD, expected loss, stress testing (~10× downturn), a scorecard master scale, and out-of-time / out-of-regime validation |
| consumer-credit-pd-ead-scorecard | PD scorecard (logistic regression + WOE/IV) with full validation, monitoring, and governance; EAD analysed and reframed as a documented data-quality finding |
| external-benchmark | A reproducible engine turning Australian bank & regulator disclosures (Pillar 3, APRA, RBA) into traceable PD / LGD / ECL / stress model inputs — with governance, an audit trail, and a 595-test suite |
| industry-analysis | Turns public ABS / RBA / PTRS data into industry risk scores, downturn / stress overlays, and macro-regime flags for commercial credit |
| mortgage-portfolio-monitoring | Loan-level mortgage monitoring on Freddie Mac data: delinquency transition / migration matrices, roll rates, IFRS 9 stage movements, an early-warning watchlist, and vintage tracking |
| commercial-portfolio-monitoring | Commercial-loan monitoring on real SBA 7(a) data: industry & state concentration (HHI, top-N), charge-off rates, vintage cohort curves, loan-age transitions, and early-warning flags |
| Together these form one stack: macro & industry overlays + external benchmarks → PD / LGD / EAD modelling → portfolio monitoring → validation. |
- CPA-qualified · Master of Accounting and Applied Finance, University of Sydney
- ~4 years as a commercial credit analyst at a Sydney private credit lender (~A$50M SME and property-backed book)
- Earlier: built and deployed logistic-regression and machine-learning models in production as a data analyst — demand forecasting, customer targeting, and reporting automation
- Google Data Analytics Professional Certificate · AML & KYC Fundamentals (AUSTRAC)
- Bilingual: English and Mandarin (including reading)