Peaks-Over-Threshold (POT)
FRM Part 2 glossary · Market Risk
Definition
EVT method modeling all exceedances above a high threshold with the GPD; data-efficient versus block maxima.
Peaks-Over-Threshold (POT) in plain English
When you're worried about rare disasters — a once-a-decade flood, a market crash — you don't have many examples to learn from, so you want to squeeze every drop of information out of the extreme events you DO have. Peaks-Over-Threshold does exactly that: instead of only looking at the single worst day each year (which throws away tons of data), it grabs every single day that blew past a high danger line and studies all of them together. It then fits a special curve called the Generalized Pareto Distribution to those "overshoots" to predict how bad the truly extreme cases could get. Think of tracking floods by recording every time the river rises above the danger mark, not just the highest flood each year — you learn far more from the same river.
Also called: POT
Learn these first
Related FRM Part 2 terms
FRM Part 2 practice questions · FRM Part 2 flashcards · FRM Part 2 study guide