Canonical: https://knowlpedia.com/probability/probability-density-function/

# Probability density function

A nonnegative measurable function whose integrals give probabilities relative to Lebesgue measure.

A **probability density function** of a [probability measure](https://knowlpedia.com/probability/probability-measure/) \(P\) on \(\mathbb R^d\) is a nonnegative measurable function \(f\) such that, for every Borel set \(A\),
\[
P(A)=\int_A f(x)\,dx,
\]
where the [integral](https://knowlpedia.com/measure-theory/lebesgue-integral/) is with respect to [Lebesgue measure](https://knowlpedia.com/measure-theory/lebesgue-measure/).

## Properties

Necessarily \(\int_{\mathbb R^d}f(x)\,dx=1\). Densities are unique only up to changes on sets of Lebesgue measure zero. The density of a random variable means a density of its [law](https://knowlpedia.com/probability/distribution-law/).

A density can exceed one; probabilities are integrals over sets. Not every probability measure has a density with respect to Lebesgue measure.

## General reference measures

Relative to another measure \(\nu\), a density is a [Radon–Nikodym derivative](https://knowlpedia.com/measure-theory/radon-nikodym-derivative/) \(dP/d\nu\). A [probability mass function](https://knowlpedia.com/probability/probability-mass-function/) is a density relative to counting measure on a countable state space.

## Examples

The [normal distribution](https://knowlpedia.com/probability/normal-distribution/) has a Gaussian density; the [uniform distribution on an interval](https://knowlpedia.com/probability/uniform-distribution/) has constant density on that interval.
