About this module
Observed data describes what happened.
Probability gives us a language for uncertain outcomes and explicitly assumed models.
Payment events, fraud flags, and claim losses will introduce conditioning, base rates, expectation, and dependence.
Watch the denominator: changing the conditioning population changes the question.
Lessons
- Lesson 51
Experiments, Outcomes, Sample Spaces, and Events
Defining what can happen to a payment
2:31 - Lesson 52
Probability Rules, Complements, Unions, and Intersections
Avoiding double-counting fraud rules
2:25 - Lesson 53
Conditional Probability
Changing the denominator after a fraud flag
2:30 - Lesson 54
Independence and Dependence
Checking whether two failure mechanisms share information
2:38 - Lesson 55
Bayes’ Theorem and Base Rates
Why a good fraud detector can still generate many false alarms
2:35 - Lesson 56
Discrete and Continuous Random Variables
Distinguishing uncertain counts from uncertain amounts
2:36 - Lesson 57
Expected Value
Budgeting expected insurance loss without promising the average
2:35 - Lesson 58
Variance, Moments, and Moment-Generating Intuition
Measuring uncertainty around an expected claim count
2:39 - Lesson 59
Joint, Marginal, and Conditional Distributions
Seeing default risk inside customer segments
2:29 - Lesson 60
Covariance and Correlation of Random Variables
Dependence in shared credit or insurance losses
2:38