Credit Risk Analytics using Python and R Training

About This Course

Credit risk remains one of the most consequential disciplines in modern banking and financial services. As regulators raise expectations around model transparency, stress testing, and capital adequacy, institutions are increasingly demanding professionals who can combine deep credit risk knowledge with practical data science capabilities. The ability to build probability-of-default models, estimate loss given default, and validate scorecards using open-source tools such as Python and R has become a core competency for risk analysts, quantitative modellers, and credit officers alike.

This 10-day programme delivered by TrainingComplex provides a rigorous, end-to-end curriculum that bridges credit risk theory and applied analytics. Beginning with the foundations of credit risk measurement and the Basel regulatory framework, the course progresses through data wrangling, exploratory analysis, logistic regression, machine learning-based credit models, and survival analysis techniques. Each day builds on the last, ensuring participants develop both conceptual clarity and the hands-on coding confidence to implement solutions in their own organisations.

Python and R are taught in tandem throughout the programme, allowing participants to appreciate the complementary strengths of each language — Python's extensive machine learning ecosystem and R's mature statistical modelling and visualisation libraries. Real banking datasets are used throughout, and participants complete a capstone project in which they design, build, and present a full credit risk model pipeline from raw data through to regulatory reporting outputs.

This course is ideal for professionals working in credit risk, retail and corporate lending, model validation, treasury, or financial regulation who wish to upgrade their analytical toolkit and contribute meaningfully to data-driven risk management. No prior Python or R experience beyond basic familiarity is assumed, though participants with some quantitative background will progress the furthest.

Who Should Attend

• Credit risk analysts and senior credit officers in banks, microfinance institutions, and development finance organisations
• Quantitative modellers and model validation professionals seeking hands-on programming skills
• Risk managers responsible for IFRS 9, Basel III/IV, or internal ratings-based (IRB) framework implementation
• Data scientists transitioning into financial services or credit risk roles
• Regulatory and compliance officers overseeing model governance and capital adequacy reporting
• Treasury and portfolio management professionals involved in loan book analysis and stress testing
• Financial technology (fintech) professionals building credit scoring engines for digital lending platforms

Learning Objectives

• Explain the conceptual and regulatory foundations of credit risk, including PD, LGD, EAD, and expected loss under Basel and IFRS 9 frameworks
• Set up and navigate Python and R environments for credit risk analytics, including key libraries such as scikit-learn, pandas, statsmodels, caret, and scorecardR
• Clean, explore, and engineer features from raw credit and loan performance datasets using both Python and R
• Build and interpret logistic regression scorecards with weight of evidence (WoE) transformation and information value (IV) analysis
• Develop, tune, and evaluate machine learning models — including decision trees, random forests, gradient boosting, and XGBoost — for credit classification tasks
• Apply survival analysis and vintage analysis techniques to model time-to-default and lifetime expected credit loss
• Estimate LGD and EAD using regression-based and model-based approaches aligned with regulatory requirements
• Conduct model validation, back-testing, and sensitivity analysis using industry-standard statistical tests and metrics
• Build stress testing scenarios and translate credit risk model outputs into capital and provision calculations
• Design a model governance and documentation framework consistent with SR 11-7, EBA guidelines, and institutional risk appetite
• Present and communicate credit risk model outputs effectively to both technical and non-technical stakeholders
• Complete an end-to-end capstone credit risk modelling project from data ingestion through to regulatory reporting deliverables

Course Modules

Requirements & Prerequisites

Participants should have a working knowledge of credit risk concepts such as loan classification, provisioning, and portfolio management. Familiarity with basic statistics — including regression, probability distributions, and hypothesis testing — is expected, as the course moves quickly into applied modelling. Prior exposure to Python or R at a beginner level is helpful; participants who have never written code before are advised to complete a short introductory programming tutorial before attending.

Training Methodology

• Guided live-coding sessions in both Python and R, with instructors walking through model-building steps on real banking datasets that participants replicate on their own laptops
• Case study analysis drawn from retail banking, SME lending, mortgage portfolios, and digital credit platforms across African and Middle Eastern markets
• Scorecard development workshops in which participants build a complete WoE/IV logistic regression scorecard from raw loan application data
• Model comparison exercises where participants benchmark logistic regression against machine learning alternatives using ROC-AUC, KS statistic, and Gini coefficient
• Group stress-testing simulation in which teams apply adverse macroeconomic scenarios to a shared loan portfolio and present capital impact findings
• Peer model review sessions modelled on industry validation practices, where participants critique each other's model assumptions and documentation
• Capstone project with structured milestones across the programme, culminating in a final presentation of a full credit risk modelling pipeline to the cohort
• Regulatory mapping exercises linking model outputs to IFRS 9 staging criteria, Basel IRB requirements, and supervisory reporting templates

Organisational Benefits

• Institutions gain staff who can develop and maintain in-house credit scoring and PD/LGD/EAD models, reducing reliance on costly external vendors
• Improved model quality and validation rigour supports regulatory compliance with Basel III/IV, IFRS 9, and central bank supervisory expectations
• Data-driven credit decisions reduce non-performing loan ratios and improve portfolio profitability
• Python and R skills enable faster, reproducible, and auditable model development workflows across the risk function
• Strengthened stress testing and scenario analysis capabilities enhance board-level risk reporting and capital planning
• Organisations build internal model governance capacity, reducing regulatory censure risk and model-related capital add-ons
• Cross-functional teams benefit from a shared analytical language, improving collaboration between credit, finance, technology, and compliance departments

Where Is Credit Risk Analytics using Python and R Training Offered?

This course runs in-person across: Abuja, Arusha, Nakuru, Addis Ababa, Dubai, Nairobi, Zanzibar, Kisumu, Pretoria, Kigali, Kampala, Cape Town, Lagos, Riyadh, Naivasha, Mombasa, Dar es Salaam, Accra, Johannesburg, plus Live Virtual (online) cohorts. Browse live dates for your city below.

Upcoming Schedules

+227 more dates — view live schedule →

Frequently Asked Questions