Lesson 1Designing target variables: default definitions, delinquency thresholds, time horizonsLearn to set target variables for credit risk models, including failure events, late payment limits, and time spans, and grasp how these choices shape model work, tuning, and business fit in lending.
Defining default and charge-off eventsChoosing delinquency and DPD thresholdsObservation and prediction window designHandling cures, roll rates, and recoveriesAligning targets with business objectivesLesson 2Customer features to collect: payment history, days past due, invoice-level data, order frequency, average invoice size, sector, size, locationSpot main customer features for credit rating, like payment past, days late, bill-level acts, order times, average bill size, industry, scale, and place, and learn to build steady, foretelling variables from raw info.
Payment history and delinquency metricsDays past due and aging bucketsInvoice-level and transaction featuresOrder frequency and seasonality signalsSector, size, and geographic attributesLesson 3Model interpretability and explainability methods (SHAP, feature importance, monotonic constraints)Learn clarity ways for credit risk models, including whole and spot feature weight, SHAP values, part reliance, and steady limits, to aid rule checks, customer tells, and inner trust in finance.
Global versus local explanationsFeature importance and permutation testsSHAP values for local explanationsPartial dependence and ICE plotsMonotonic constraints in credit modelsLesson 4Augmented features: external data (credit bureau, public filings), macro indicators, trade supplier behaviorLook into adding to inner records with bureau info, public papers, big economy signs, and trade supplier acts to make strong credit features that boost model truth, steadiness, and early alerts over cycles.
Using credit bureau scores and tradelinesIncorporating public filings and legal eventsMacroeconomic indicators for credit cyclesTrade credit and supplier payment behaviorData quality checks and reconciliationLesson 5Bias and risk mitigation: label bias, sample selection, proxy variables, data imbalance and mitigation strategies (regularization, fairness checks)Check bias sources in credit data, like label tilt, sample pick, stand-in variables, and class unbalance, and learn cut ways including reweight, steadying, fairness checks, and careful feature and rule design.
Label bias and historical decision effectsSample selection and survivorship biasProxy variables and sensitive attributesClass imbalance and resampling methodsFairness metrics and monitoring checksLesson 6Common predictive models for credit risk: logistic regression, random forest, gradient boosting (LightGBM/XGBoost)Go over usual foretelling models for credit risk, like logistic line, random trees, and boost methods such as XGBoost and LightGBM, comparing strengths, limits, tuning ways, and rollout thoughts for finance.
Logistic regression for PD estimationTree-based models and random forestsGradient boosting, XGBoost, LightGBMHyperparameter tuning and validationModel comparison and champion–challengerLesson 7Model governance: validation, backtesting, calibration, population stability, and performance monitoringGrasp model oversight for credit risk, including free checks, past tests, tuning, group steadiness, and steady watch, to meet rule needs and keep trusty live models in financial operations.
Independent model validation scopeBacktesting and benchmarking approachesCalibration and probability of default testsPopulation stability and drift metricsOngoing performance dashboards and alertsLesson 8Decisioning: converting model outputs to risk tiers, credit limits, payment terms and approval workflowsFind how to turn model results into working credit choices, like risk levels, credit caps, payment times, and okay flows, balancing risk taste, customer feel, and portfolio gain in lending.
Score-to-risk tier mapping and cutoffsSetting credit limits from PD and LGDDesigning payment terms by risk bandApproval workflows and escalation rulesChampion–challenger decision strategies
