PH
Enterprise Risk Management In Retail Banks And Credit Unions
Published 8/2026
MP4 | Video: h264, 1920x1080 | Audio: AAC, 44.1 KHz, 2 Ch
Language: English | Duration: 4h 8m | Size: 1.62 GB
From zero risk experience to examiner-ready models. 25 modules, every script downloadable and reproducible.
What you'll learn
Build a probability-of-default scorecard end to end, then assemble PD, LGD and EAD into a working expected-loss engine
Calculate a CECL allowance three ways and project capital under DFAST-style and NCUA-style stress scenarios
Model interest rate and liquidity risk - repricing gap, net interest income, economic value of equity, deposit decay curves, and stress survival horizon
Build fraud and BSA/AML detection models, run fair lending and CRA analysis, and pull every risk domain into an executive dashboard
Requirements
Basic Python - variables, functions, and a willingness to read a pandas DataFrame No prior risk, banking, or credit union experience of any kind. The course builds it from lecture one No prior regulatory knowledge. Regulations are taught where they apply, not as a reading list An ordinary laptop. Everything runs locally - no cloud account, no ρáíd software, no institutional data access Setup is covered by short standalone modules, so you can arrive with nothing installed
Description
This course contains the use of artificial intelligence.
Most risk training gives you vocabulary. This course gives you the work.
By the last lecture you will have built a probability-of-default scorecard, a full expected-loss engine, a CECL allowance three different ways, an interest rate risk model, a liquidity stress test, a fraud detector, a BSA/AML monitoring system, a capital stress test, and an executive risk dashboard - in Python, on your own machine, with every script downloadable and yours to keep.
Not slides about code. Code.
How the course works
Every analytic lecture is a code-along. You download the script, open it beside the video, and run it as the instructor runs it - same fixed seed, same numbers, same charts on your screen as on theirs. Nothing is typed under time pressure and nothing is left as an exercise you cannot check. When a model gives an odd answer, you see the odd answer too, and you find out why.
Conceptual lectures - governance, regulation, leadership - use slides, because there is nothing to run. Everything else puts the code on screen.
Banks and credit unions, side by side
Almost every course picks one and leaves you to translate. This one teaches both frameworks together: OCC, Federal Reserve, and FDIC alongside NCUA, throughout. You will stress net worth under Part 702 and risk-based capital under the bank rules, underwrite member business loans and commercial credit, and file a 5300 and a Call Report. If you move between the two sides of the industry, and many risk careers do, you will not be starting over.
Regulation taught where it bites
No reading lists. SR 11-7 appears when you validate a model. ASC 326 appears when you build the allowance. The Bank Secrecy Act appears when you risk-rate a member and tune a monitoring scenario. Reg E, Reg DD, ECOA, and UDAAP appear when you price a product, resolve a dispute, or monitor an override. You will finish knowing what an examiner asks for, what documentation satisfies it, and what an undocumented assumption costs you in an exam.
The part that gets people promoted
The course also teaches where the work sits in the organization: the three lines of defense, and how the same regression carries different accountability depending on whose risk it informs. Second-line independent measurement and challenge is a different job from first-line analytics that prices loans and sets cutoffs, and knowing which hat you are wearing is treated here as a professional skill, not a footnote. That distinction is most of the difference between an analyst and a director.
What you get
25 modules and 240+ lectures, ten to twenty-five minutes each
A downloadable, deterministic Python script with every analytic lecture
Starter exercises and a fully worked answer key, with the common mistakes annotated rather than hidden
A five-question knowledge check per lecture, with explanations
A twenty-question quiz bank at the end of each section
Five optional capstone projects that integrate the whole course
Standalone setup modules, so you can arrive with nothing installed and still run the first script alongside the instructor
Nothing proprietary, everything reproducible
Every dataset is either generated inside the course with a fixed seed or pulled from public FFIEC and NCUA sources. No confidential data, no results you had to be there to see. Every number you see, you can reproduce, which also means every script is yours to adapt and take to work on Monday.
Start from zero. Finish expert.
Zero prior exposure to enterprise risk management is required. Not the vocabulary, not the regulations, not the frameworks. The course builds all of it from first principles.
That is a statement about where the course starts, not where it stops. The aim is comprehensive coverage of every ERM topic that matters in a US retail bank or credit union, at the depth a practitioner actually works at: the assumptions behind each method, where it breaks, what an examiner challenges, and what you do when the standard approach does not fit.
Retain what is taught here and you are not conversant in enterprise risk management. You are expert in it.
What you will learn
Build a probability-of-default scorecard end to end - weight of evidence, monotonic binning, logistic regression, and KS/Gini/AUC/PSI validation
Assemble a full expected-loss engine from PD, LGD, and EAD models, then reuse it for CECL and capital stress testing
Calculate a CECL allowance three ways - vintage chain-ladder, roll-rate migration, and discounted lifetime PD x LGD x EAD with macro scenarios
Model interest rate risk with repricing gap, net interest income simulation, economic value of equity, deposit betas, and rate shocks
Run liquidity analytics - deposit decay curves, surge deposits, depositor concentration, stress survival horizon, and a contingency funding plan
Detect fraud with XGB oost on imbalanced data, anomaly detection, network linkage analysis, and SHAP reason codes
Tune a BSA/AML transaction monitoring system with below-the-line testing, and cut OFAC screening false positives
Test for fair lending, redlining, steering, and UDAAP exposure, and run a CRA distribution analysis
Price a loan from cost of funds, expected loss, and capital, and evaluate it on RAROC
Project capital under DFAST-style and NCUA-style stress scenarios with macro satellite models and PPNR
Produce Call Report and NCUA 5300 outputs with edit checks, plus board-ready risk packages
Build an executive risk dashboard that brings every risk domain into a single view
Requirements
Basic Python - variables, functions, and a willingness to read a pandas DataFrame
No prior risk, banking, or credit union experience of any kind. The course builds it from lecture one
No prior regulatory knowledge. Regulations are taught where they apply, not as a reading list
An ordinary laptop. Everything runs locally - no cloud account, no ρáíd software, no institutional data access
Setup is covered by short standalone modules, so you can arrive with nothing installed
Who this course is for
Risk analysts and managers at banks and credit unions who want to move from reporting into modeling Data analysts moving into financial services who know Python but not the regulatory frame Finance, treasury, lending, and compliance professionals who need to read, challenge, and use risk models credibly Model validation and internal audit staff who have to review this work and want to build it once themselves Experienced practitioners preparing for senior and director-level roles, where the job is building and leading the analytics function rather than running a single model Anyone who has been handed a risk model they did not build and could not defend Not for you if you want a certification cram course, or theory without code. Every module expects you to run something