KomFi. Academy

Heteroskedasticity

FRM Part 1 glossary · Quantitative Analysis

Definition

The condition in which the variance of a regression's error term is not constant across observations, violating a classical assumption and rendering ordinary standard errors unreliable.

Heteroskedasticity in plain English

When you fit a straight line through a cloud of data points to predict something, the line is never perfect — each point misses by some amount, called the "error." Heteroskedasticity just means those misses are small in some parts of the data and big in others, instead of being roughly the same size everywhere. Picture predicting people's spending from their income: poor households are all forced to spend on about the same basics (small, tidy misses), but rich households swing wildly from thrifty to lavish (huge misses) — so the scatter fans out. This matters because the usual formulas that tell you "how confident should I be in this line?" secretly assume the misses are equally sized, so when they aren't, your confidence numbers are wrong and you can fool yourself into trusting a shaky result.

Learn these first

Sign up free: all 170 FRM Part 1 terms as flashcards, plus practice questions with full explanations →

Related FRM Part 1 terms

FRM Part 1 practice questions · FRM Part 1 flashcards · FRM Part 1 study guide