Home › Glossary › Data › Empirical cumulative distribution function (eCDF or EDF)

Intermediate · Data

Empirical cumulative distribution function (eCDF or EDF)

Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.

TL;DR. A function showing the proportion of data points at or below a specific value, based on observed data.

Technical Definition

A function showing the proportion of data points at or below a specific value, based on observed data.

How it works

An empirical cumulative distribution function (eCDF) plots the proportion of data observations that fall at or below a given value. Unlike a theoretical CDF, it is derived directly from a sample of data, providing a direct visualization of the observed distribution. It is useful for understanding data spread and comparing distributions visually.

Related Concepts

  • Cumulative distribution function (CDF) — A CDF describes the probability that a random variable will take a value less than or equal to a specific point.

Further Reading

  • Google ML Glossary