Canonical: https://knowlpedia.com/probability/cumulative-distribution-function/

# Cumulative distribution function

The function assigning to each real threshold the probability of a value at or below it.

The **cumulative distribution function** of a [probability measure](https://knowlpedia.com/probability/probability-measure/) \(P\) on \(\mathbb R\) is
\[
F:\mathbb R\to[0,1],\qquad F(t)=P(({-\infty},t]).
\]

## Random variables

For a real-valued [random variable](https://knowlpedia.com/probability/random-variable/) \(X\), this is the cumulative distribution function of its [law](https://knowlpedia.com/probability/distribution-law/): \(F_X(t)=\mathbb P(X\le t)\).

## Properties

The function is nondecreasing and right-continuous, with limits zero at \(-\infty\) and one at \(+\infty\). It determines the measure through \(P((a,b])=F(b)-F(a)\) for \(a<b\).

## Masses and densities

For a [mass function](https://knowlpedia.com/probability/probability-mass-function/) on a countable subset of \(\mathbb R\), \(F(t)=\sum_{x\le t}p(x)\). If \(P\) has a [density](https://knowlpedia.com/probability/probability-density-function/) \(f\), then \(F(t)=\int_{-\infty}^t f(x)\,dx\). A cumulative distribution function exists even when neither representation applies.
