Canonical: https://knowlpedia.com/probability/chebyshev-inequality/

# Chebyshev's inequality

Upper bound on deviation probability using variance.

**Chebyshev's inequality:** Let \(X\) be a [random variable](https://knowlpedia.com/probability/random-variable/) with [expectation](https://knowlpedia.com/probability/expectation/) \(\mu=\mathbb{E}[X]\) and finite [variance](https://knowlpedia.com/probability/variance/) \(\sigma^2=\mathrm{Var}(X)\). Then for every \(t>0\),
\[
\mathbb{P}\bigl(|X-\mu|\ge t\bigr)\le \frac{\sigma^2}{t^2}.
\]

## Equivalent characterizations

Equivalently, for every \(k>0\),
\[
\mathbb{P}\bigl(|X-\mu|\ge k\sigma\bigr)\le \frac{1}{k^2}.
\]

## Remarks

Here \(\mathbb{P}\) denotes the [probability measure](https://knowlpedia.com/probability/probability-measure/) on the underlying [probability space](https://knowlpedia.com/probability/probability-space/). Chebyshev's inequality is a direct consequence of [Markov's inequality](https://knowlpedia.com/probability/markov-inequality/) applied to \((X-\mu)^2\), and it is a standard tool for proving the [weak law of large numbers](https://knowlpedia.com/probability/weak-law-large-numbers/).
