Definition:Median of Continuous Random Variable

From ProofWiki
Jump to navigation Jump to search

This page is about the median of a continuous random variable. For other uses, see median.

Definition

Let $X$ be a continuous random variable on a probability space $\struct {\Omega, \Sigma, \Pr}$.

Let $X$ have probability density function $f_X$.

A median of $X$ is defined as a real number $m_X$ such that:

$\ds \map \Pr {X < m_X} = \int_{-\infty}^{m_X} \map {f_X} x \rd x = \frac 1 2$

That is, $m_X$ is the first $2$-quantile of $X$.

Hence it is also the $50$th percentile of $X$.


Warning

Care should be directed towards the uniqueness of medians.

For example, if $\map {f_X} x = 0$ on some closed real interval $\closedint a b$ of $x$, we may have $\ds \int_{-\infty}^{m_X} \map {f_X} x \rd x = \frac 1 2$ for all $m_X \in \closedint a b$.


Also see

  • Results about medians can be found here.


Sources

Retrieved from "https://proofwiki.org/w/index.php?title=Definition:Median_of_Continuous_Random_Variable&oldid=733657"