Bank-style Credit Risk Scorecard using Logistic Regression, IFRS-9 Expected Credit Loss, and an Interactive Streamlit Risk Dashboard for loan default prediction.

R-codebase for a scientific research article, titled "Approaches for modelling the term-structure of default risk under IFRS 9: A tutorial using discrete-time survival analysis"

Practitioner-grade IFRS 9 expected-credit-loss engine: PD term structures, three-stage SICR waterfall, EBA macro overlays and a Vasicek/ASRF point-in-time PD. Parallel Python + Excel with a methodology deck.

End-to-End IFRS 9 LGD Model development by Advance Workout Model.

Streamlit app that computes per-loan Expected Loss, Lifetime ECL, and Risk Rating from EAD/PD/LGD/WAL Excel portfolios

This model estimates the 12-month Probability of Default (PD) for prime residential mortgage customers in the United Kingdom, aligned with the IFRS 9 impairment framework and calibrated to an adverse macroeconomic scenario. Version 1 (v1) is developed using gradient-boosted decision trees (GBDT)

End-to-end credit risk modeling in Python: WoE/IV feature engineering, logistic regression PD scorecard with points scaling, XGBoost challenger, model validation suite (KS, Gini, AUROC, PSI, calibration), SHAP explainability, and IFRS 9 three-stage expected credit loss (ECL) — on real Lending Club data, with a Streamlit dashboard.

End-to-End IFRS 9 PD Model development by Empirical Migration Matrix and Credit Cycle Index Models.

End-to-End IFRS 9 PD Model development by Cohort Model.

Interactive WOE (Weight of Evidence) and IV (Information Value) binning web tool for credit risk scoring, segmentation, and transparent scorecard development.

Distributed regression evaluation engine built with FastAPI, Celery and Redis for large-scale macroeconomic model validation and statistical analysis.

End-to-end PD/LGD/EAD credit risk platform reflecting industry-standard model development workflows across Canadian financial institutions and credit unions. Built to OSFI E-23 / IFRS 9 standards.

IFRS 9 SICR model Mortgages

IFRS 9 expected credit loss engine on 1.35M real loans: PD, LGD, EAD, three-stage staging, and probability-weighted macroeconomic scenarios

IFRS 9 Expected Credit Loss model on a 10,000-loan UK SME portfolio - PD/LGD/EAD, SICR staging, 3-scenario macro overlay. R, Python, Excel. AUC 0.77.

End-to-end credit risk and fraud risk modeling: PD, LGD, scorecards, IFRS 9 ECL, transaction fraud, and fraud ring detection.

IFRS 9 / AASB 9 mortgage credit-risk suite on Freddie Mac loan-level data — PD (logistic, AUC 0.81), real LGD from actual loss data (reconciled to the vendor loss field at 0.99), EAD, expected credit loss, stress testing (~10× downturn), a scorecard master scale, and out-of-time / out-of-regime validation.

WTI–FX hedge program: budget allocation, covariance-aware delta, IFRS 9 CFH, ESG DiD (working papers + Python + VBA)

SQL & Python project for Internal Audit testing of capital, liquidity and IFRS 9 controls in a simulated banking environment.