What you will learn in this course:

• Business Scorecards: Acquisition and Behaviour Scorecards
• BASEL AIRB Models: Probability of Default (PD), Loss Given Default (LGD) and Exposure at Default (EAD) models.
• Provision Models: Allowance for Loans and Lease Loss Models using Expected Loss Approach
• Tools: Python
• Machine Learning Algorithm: Decision Tree, Random Forest, XGBoost, Support Vector Machine
• Basic descriptive analysis
• Basic statistical interference
• Basic predictive modeling techniques
• Python Programming
• Regulatory Risk Model development and Validation
• Understanding of scorecard model vocabulary
• Detailed training on scorecard model development
• Credit risk regulatory guidelines with BASEL II
• Regulatory stress testing guidelines with DFAST and CCAR

Credit Risk Modeling at DexLab Analytics has undergone a complete industrial upgradation. The new certification module, called Credit Risk Modeling with Machine Learning, is now inclusive of latest industry trends and implementation. Year 2020 is being touted as the year of Machine Learning implementation into risk analytics. As an organization we strive towards betterment and making you industry ready.

Why This Course Exists

The financial services industry is facing a skills shortage. Risk modelers are in high demand—700+ open positions on Glassdoor India right now. Yet most training programs teach theory disconnected from practice.

You can’t learn credit risk from textbooks alone. You learn it by:
– Building Probability of Default (PD) models from real historical data
– Understanding Loss Given Default (LGD) and Exposure at Default (EAD) through actual collateral examples
– Validating models using KS statistic, GINI coefficient, and ROC curves
– Deploying machine learning (Random Forest, XGBoost) at scale
– Passing regulatory audits

That’s what this course delivers. 110 hours of intensive, hands-on training. Real bank use cases. Python +SAS+ Excel. Your models become portfolio pieces that you can show employers.

What You’ll Learn (5 Core Outcomes)

• Calculate Probability of Default (PD) from historical data: Using logistic regression in both Python and Excel. Understand how to classify borrowers by default likelihood. Learn from real datasets (anonymized HDFC, ICICI examples).
• Understand Loss Given Default (LGD) and Exposure at Default (EAD): With concrete examples from secured vs. unsecured lending. See how collateral, recovery rates, and guarantee structures impact LGD. Model EAD for revolving credit (credit cards) vs. term loans.
• Build Expected Loss models (PD × LGD × EAD): Show the balance sheet impact. Calculate capital requirements under Basel III. Understand how single models drive portfolio-wide decisions.
• Validate models using regulatory standards: KS statistic (should be top 4 deciles). GINI coefficient. ROC curves. Ensure IFRS 9 and Basel III compliance. Know when to redevelop vs. recalibrate.
• Implement machine learning for automated risk classification: Beyond logistic regression. Random Forest for non-linear patterns. XGBoost for production models. Feature engineering using IV/WOE analysis. Hyperparameter tuning.

Real-World Examples: Why This Matters

IL&FS Crisis (2018, India): An NBFC (Non-Banking Financial Company) defaulted on INR90,000 crores. The warning signs were there—stage migration indicators, liquidity stress, maturity mismatches. But risk models at connected institutions didn’t catch the contagion soon enough. A properly trained risk team using PD/LGD/EAD models would have detected the exposure concentration and advised earlier.

COVID-19 Stress (2020-21): Banks granted moratoriums on 40% of retail portfolios. ECL (Expected Credit Loss) under IFRS 9 spiked 200-300%. Risk models that incorporated macro scenarios (GDP growth, unemployment, interest rates) predicted this wave. Teams trained in forward-looking PD modeling implemented better reserve policies faster.

Archegos Capital (2021): A hedge fund took INR50,000+ crore losses due to concentrated positions in single stocks. Traditional risk models missed correlation breakdowns. Firms that trained teams in advanced ML and stress testing (like DexLab’s curriculum) detected similar risks faster.

Who This Course Is For

• Credit Analysts at national and global banks
• Risk Professionals in NBFC, HFC, and fintech companies
• Finance professionals seeking promotion into risk teams
• MBA students specializing in Finance or Risk Management
• FRM (Financial Risk Manager) candidates needing practical foundation
• Data Analysts wanting to transition into risk analytics

Why DexLab vs. Other Options

Dexlab Analytics has more than a decade of experience providing training on risk analytics.
DexLab teaches Python automation—you learn how to scale models from 1,000 to 1 million borrowers.
DexLab assumes you start from basics and teaches credit risk specifically.
DexLab is integrated, practical, project-based. You walk out with a portfolio piece.
DexLab is ₹41,999—premium quality at affordable pricing.
DexLab brings fresh perspectives, latest ML techniques, and external credibility that helps with promotions.

Salary Impact & Career Path

The data is clear. Credit risk professionals earn well in India:

Post-Training Salary Jump: Alumni report average salary increase of INR2L – INR5L within 12 months of completing this certification. That’s a 100-250x return on your INR41,999 investment in Year 1 alone.

Career Progression: Risk modelers become Risk Managers, then Risk Heads. The path to leadership in financial institutions runs through risk. Banks prioritize risk professionals for executive roles because risk management is existential—get it wrong and the institution fails.

Meet Your Instructor

Your instructor brings over 10+ years of hands-on experience in Credit Risk, Analytics, and Predictive Modeling. Previous worked at Moody’s, GE Capital, Standard Chartered and SBI — four of the world’s most risk-sophisticated institutions.

Currently working with a leading global bank, he has extensive expertise in:
– Scorecard development and regulatory validation
– Model deployment in production environments
– Advanced machine learning techniques using Python (Random Forest, XGBoost, Neural Networks)
– Basel III and IFRS 9 compliance

This isn’t theory from textbooks. This is battle-tested knowledge from someone who built models that approve or deny millions of rupees in credit every single day.

How the Course Works

Weekend Batches – Saturday & Sunday
Live Online Instruction with video, screen sharing, and real-time Q&A
Use-case Project-Based: Build real models with actual datasets (anonymized)
Peer Learning: Learn alongside 10-15 professionals from banks and fintech
Hands-On Tools: Python (Jupyter Notebooks), Excel, SQL, real banking data
Certificate of Completion: Recognized by banks for professional development

Timeline: 40 weekend sessions = 110 hours total. You can attend while working your day job.

Frequently Asked Questions

• Do I need banking experience? You need basic statistics and probability (covered in Module 1). We teach fundamentals → advanced. Prior credit risk knowledge is advantageous but not mandatory.
• Will I actually get a job after this? 300+ DexLab alumni are employed in risk roles at Wells fargo, Citi, RBS, EXL and other banks. Career support included: resume review, interview prep, job board access with 30+ active postings.
• Should I learn Python or use Excel? Excel for quick analysis and reporting. Python for automation and scaling models. We teach both—you’ll see why each has its place.
• Is this certificate recognized by banks? Banks value practical model-building skills over generic certifications. Your portfolio of models (which you build during the course) is what gets you hired.
• How is this different from free online tutorials? Free tutorials are self-paced and fragmented. You get stuck, there’s no one to help. Here: Live instructor, real-time Q&A, peer learning, structured curriculum, accountability, mentoring.
• Can I use the models immediately at my current job? Bring your bank’s data (anonymized), build models in the course, and deploy them at work. Many students have done exactly this.
• How much time do I need to commit outside of classes? Plan for 4-5 hours per week for assignments and projects (beyond the 4 hours in class). Real learning requires practice.

Ready to Master Credit Risk Modeling?

Enrollment for Next Batch: Sunday, 12th July 2026
Limited Seats Available: Maximum 10 participants per batch to ensure quality
Price: ₹41,999 + GST (18%)

Enroll Now: Click Here
Download Full Syllabus: Click Here

Questions? Contact us:
Email: hello@dexlabanalytics.com
Phone: +91 9903662244

What is Credit Risk?

Credit risk is the potential loss from a borrower’s failure to repay a loan or meet contractual obligations. In Indian banking, it represents the largest risk category affecting bank profitability. The Reserve Bank of India (RBI) estimates that credit losses account for 60-70% of total bank losses annually. This is why understanding and modeling credit risk is critical for every bank.
Credit risk has three main components that work together PD, LGD and EAD. These three combine in a single formula: Expected Loss = PD × LGD × EAD

Probability of Default(PD)

Probability of Default (PD) measures the likelihood that a borrower will fail to repay within a specific time period, typically one year. PD is expressed as a percentage ranging from 0% to 100%. PD is the first and most critical input in calculating expected losses. Under Basel III regulations, banks must estimate PD accurately for capital adequacy requirements. Regulatory capital requirements depend directly on PD estimates. A 1% change in PD can significantly impact capital requirements and bank profitability.

Loss Given Default (LGD)

Loss Given Default (LGD) measures the percentage of exposure lost when a borrower defaults. LGD is calculated using the formula:
LGD = (Exposure – Recovery) / Exposure

Exposure at Default (EAD)

Exposure at Default (EAD) is the total value at risk when a borrower defaults. EAD represents the actual amount of money the bank has lent out when the default occurs.
For Term Loans (home loans, auto loans, corporate loans):
EAD = Outstanding loan amount at the time of default
For Revolving Credit (credit cards, lines of credit):
EAD = Outstanding balance + Undrawn credit line

Portfolio Credit Risk

Portfolio credit risk is the aggregate risk from all loans in a bank’s portfolio combined. While individual loan risk is calculated as PD × LGD × EAD, portfolio risk considers how losses correlate across thousands of loans.

Key Portfolio Metrics:

Expected Loss (EL): The average expected loss across all loans in the portfolio. This is predictable and banks provision for this.

Unexpected Loss (UL): Potential loss above the expected level. This determines minimum capital requirements.

Value-at-Risk (VaR): The maximum loss the bank could face at a given confidence level, typically 99%. VaR at 99% means there’s only a 1% chance of losses exceeding this amount.

Regulatory Framework (Basel III & IASB)

Credit risk modeling is governed by two major international regulatory frameworks: Basel III for capital adequacy and IASB (International Accounting Standards Board) for financial reporting. Indian banks must comply with both frameworks as set by the Reserve Bank of India (RBI).

Basel III Framework:

Basel III was developed by the Basel Committee on Banking Supervision following the 2008 financial crisis to strengthen bank capital requirements. It sets minimum capital standards that all major banks globally must follow.

The Core Basel III Requirement: Capital Ratio = Regulatory Capital / Risk-Weighted Assets

IASB Framework (IFRS 9):

The International Accounting Standards Board (IASB) sets financial reporting standards through IFRS (International Financial Reporting Standards). IFRS 9, which replaced IAS 39, fundamentally changed how banks account for credit losses.