Limit of Cumulative Distribution Function at Negative Infinity

Theorem

Let $\struct {\Omega, \Sigma, \Pr}$ be a probability space.

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

Then:

$\ds \lim_{x \mathop \to -\infty} \map {F_X} x = 0$

where $\ds \lim_{x \mathop \to -\infty} \map {F_X} x$ denotes the limit at $-\infty$ of $F_X$.

for all decreasing real sequences $\sequence {x_n}_{n \mathop \in \N}$ with $x_n \to -\infty$ we have $\map {F_X} {x_n} \to 0$

at which point we will obtain:

$\ds \lim_{x \mathop \to -\infty} \map {F_X} x = 0$

Since $\sequence {x_n}_{n \mathop \in \N}$ is decreasing, we have:

the sequence $\sequence {\hointl {-\infty} {x_n} }_{n \mathop \in \N}$ is decreasing.

$\ds \bigcap_{n \mathop = 1}^\infty \hointl {-\infty} {x_n} = \O$

Let $P_X$ be the probability distribution of $X$.

$\ds \map {P_X} \O = \lim_{n \mathop \to \infty} \map {P_X} {\hointl {-\infty} {x_n} }$

So, we obtain:

$\ds 0$

$=$

$\ds \map {P_X} \O$

$\ds $

$=$

$\ds \lim_{n \mathop \to \infty} \map {P_X} {\hointl {-\infty} {x_n} }$

$\ds $

$=$

$\ds \lim_{n \mathop \to \infty} \map {F_X} {x_n}$

Since $\sequence {x_n}_{n \mathop \in \N}$ was arbitrary, we have:

for all decreasing real sequences $\sequence {x_n}_{n \mathop \in \N}$ with $x_n \to -\infty$ we have $\map {F_X} {x_n} \to 0$.

$\ds \lim_{x \mathop \to \infty} \map {F_X} x = 0$

$\blacksquare$