Credit losses are the largest threat to bank and non-bank lender profitability, and rising competition from digital lenders has made speed and accuracy in credit decisions a commercial necessity. Supervisors and auditors expect forward-looking provisioning and documented, validated models. Institutions that cannot explain what drives their defaults struggle to price risk or to grow their loan books safely.

Data and Analytics for Credit Risk & Lending Training Course teaches the full analytical chain behind modern lending. Participants start with data quality, definitions of default and sampling, then explore variables and build credit scorecards, from binning and weight of evidence to model fitting and validation. The course explains how probability of default, loss given default and exposure at default combine into expected credit loss under IFRS 9, and how stage allocation and forward-looking information affect provisions. Portfolio analytics follow, covering concentration, migration matrices, vintage curves, collections performance and early warning systems.

Delivery is based on practice. Participants run each technique on sample loan data, interpret outputs critically, discuss model risk and fairness, and prepare a short portfolio quality report suitable for a credit committee.

By the end of the course, participants will be able to:

• Prepare and validate lending data and define default consistently.
• Engineer and select predictive variables for credit models.
• Develop and validate application scorecards using logistic regression.
• Interpret discrimination and stability measures such as Gini, KS and PSI.
• Estimate probability of default, loss given default and exposure at default.
• Calculate IFRS 9 expected credit loss with staging and forward-looking overlays.
• Analyse portfolio quality using vintage, roll-rate and migration methods.
• Design early warning indicators and collections strategies from data.

The programme is designed for professionals who make, model or oversee lending decisions, including:

• Credit analysts and underwriters
• Credit risk managers and portfolio managers
• Retail, SME and corporate lending heads
• Microfinance institution and SACCO managers
• Model developers and model validators
• Internal auditors and risk compliance officers
• Fintech and digital lending product teams
• Finance staff responsible for loan loss provisioning

Participants leave with:

• A working scorecard built and tested on a sample loan dataset.
• A reusable expected credit loss calculation workbook.
• Templates for vintage and roll-rate analysis.
• A short list of early warning indicators suited to their portfolio.
• A model validation checklist for challenging internal or vendor models.
• A draft portfolio quality report layout for credit committees.

The course follows the model-building process in the order lenders actually use it. Participants work through:

• Guided labs on anonymised retail, SME and microfinance loan data
• Side-by-side Excel and Python walk-throughs of key calculations
• Critique sessions on model outputs and validation results
• Group discussion of provisioning judgements and overlays
• A closing exercise preparing a portfolio quality summary

Participants who attend all sessions and finish the lab exercises are awarded a CPD-accredited Certificate of Completion by Vision Reach Global Consultancy.