Zoom Training Details:

Day 1:

Duration	Topic
80 min	Python, Jupyter and Spyder Efficient presentations in Jupyter Notebook using RISE  extension Key concepts in Python using pandas, numpy, statsmodels and scikit-learn Outcome analysis: loan amortization and default Impact of COVID-19, Rates/Expenses/Climate shocks on Credit Risk COVID-19 challenges Interest and expense increases Climate change challenges
20 min	Break
80 min	Machine Learning Concepts for Probabilities of Default (PDs) Statistical and machine learning thinking Workflow in machine learning Loss functions Regularisation penalty Learning rate Train-test validation split Standardisation and scaling Bias-variance trade-off Cross-validation Hyper-parameters and tuning Supervised and unsupervised learning Classification and regression vs. dimensionality reduction

Day 2:

Duration	Topic
80 min	Machine Learning Validation for PDs Validation framework Backtesting PDs: discrimination, calibration, stability Confusion matrix, sensitivity, specificity, accuracy, ROC curves, AUC & accuracy ratio Brier scores Binomial test Jeffreys prior test Hosmer-Lemeshow Calibration curve and reliability diagram Tests on model stability Portfolio dependence Traffic lights Qualitative validation
20 min	Break
80 min	Statistical Concepts for PDs Outcome Engineering: time-vintage-age (TVA) analysis Default definitions Non-monotone relations between defaults and risk factors Weight-of-evidence, categorisation, splines Discrete time models (Logit & Probit) Comprehensive modelling including vintage, age and time effects Maximum likelihood estimation Significance testing Hands-on Excercises in Python: Visualisation and Reporting through Dashboards Communication with non-Python users User interfaces

Day 3:

Duration	Topic
80 min	Unsupervised Machine Learning for PDs K Means clustering Hierarchical clustering Principal component analysis Supervised Machine Learning for PDs Logistic regression K nearest neighbours Decision trees Hyperparameter tuning
20 min	Break
80 min	Crisis PDs Asymptotic Single Risk Factor Model Asset correlations Basel calibrations Through-the-cycle & point-in-time Predicting financial crises Stress-testing for economic downturns and climate risk Low default portfolios and margin of Conservatism (MoC)

Day 4:

Duration	Topic
80 min	Supervised Machine Learning for PDs Bagging and boosting Random forests Light Gradient Boosting Machines (GBM) Extreme Gradient Boosting Machines (XGBM) Neural networks Network architecture Activation function Deep learning Do it like Google: Introduction into Keras/Tenserflow
20 min	Break
80 min	Payoff probabilities Discrete time models (Logit & Probit) Selection models Multi-period PDs and Lifetime PDs for CECL and IFRS 9 Roll rate analysis Loan loss provisioning and Basel capital 12-month expected loss vs. lifetime expected loss Rating class formation and roll rate analysis Macroeconomic forecasts Multi-period PD forecasts based on the macroeconomy and lifecycle Prediction of lifetime expected losses Significant increase in credit risk (SICR)

---

Back to Live