PRM study plan

Define operational resilience and distinguish it from traditional operational risk so you can state the difference in two sentences without notes

Map the cause-impact-outcome chain across the four resilience pillars (critical operations, impact tolerances, mapping, testing) and label each link with one concrete banking example

List the five most common single points of failure in payment and settlement chains and explain why each is hard to substitute under stress

Draft a one-page glossary of 15 resilience terms (impact tolerance, severe but plausible scenario, recovery time objective, business service) with a one-line definition each

Build a severe-but-plausible scenario for a third-party cloud outage and identify the impact tolerance breach point for retail payments

Rank five recovery and continuity options (redundancy, geographic diversification, manual fallback, contractual exit, mutual aid) by cost, speed, and residual risk

Explain how concentration risk in a small number of critical technology vendors amplifies operational resilience exposure

Summarize the supervisory expectations for self-assessment, scenario testing, and board reporting in your own words

Distinguish supervised from unsupervised learning and match each to a concrete risk use case (fraud scoring versus customer segmentation)

Explain bias-variance tradeoff and overfitting using a credit-default example, including how regularization and cross-validation address them

Walk through how a decision tree splits on information gain and why ensembles such as random forests and gradient boosting usually outperform a single tree

Describe precision, recall, F1, and AUC-ROC and state which metric you would prioritize for rare-event fraud detection and why

Diagnose model risk sources: data leakage, sample selection bias, unstable features, and concept drift over a credit cycle

Outline a validation workflow covering data quality checks, out-of-time testing, benchmarking, and sensitivity analysis

Explain explainability tools (feature importance, SHAP, partial dependence) and when a black-box model is defensible

Describe governance requirements for model inventory, tiering, periodic review, and challenger models

Differentiate fintech business models (payments, lending, robo-advisory, regtech, distributed ledger) and name the dominant risk in each

Explain open banking and API-driven data sharing, including consent, liability, and data-security implications

Compare tokenization, biometric authentication, and behavioral analytics as fraud controls, noting failure modes for each

Assess how digital onboarding and alternative data change KYC, AML, and credit underwriting judgments

Trace how a digital lending platform originates, funds, and services a loan, and identify where credit, liquidity, and operational risk concentrate

Explain stablecoin and crypto-asset exposure risks including custody, liquidity mismatch, and settlement finality

Summarize the core Basel framework elements: capital tiers, risk-weighted assets, leverage ratio, and liquidity coverage ratio

Connect fintech disruption to traditional risk topics by describing how outsourcing and concentration risk appear in a bank-fintech partnership

Compute expected loss as PD times LGD times EAD and apply it to a worked portfolio example

Explain market risk measures VaR and expected shortfall, including their limitations and backtesting failure modes

Describe liquidity risk metrics (LCR, NSFR) and how a funding run propagates through a balance sheet

Outline stress testing design: scenario selection, transmission channels, and reverse stress testing

Synthesize cross-domain links by writing how a machine-learning credit model failure becomes an operational resilience event

Complete two timed 40-question mixed-domain practice sets and log every miss by domain and error type

Re-drill the 20 weakest concepts identified in practice logs until recall is fluent

Recite the key definitions and formulas for all four domains from memory in a 15-minute self-audit