Definition:Null Set

Definition

Let $\left({X, \Sigma, \mu}\right)$ be a measure space.

A set $N \in \Sigma$ is called a ($\mu$-)null set iff $\mu \left({N}\right) = 0$.

Family of Null Sets

The family of $\mu$-null sets, $\left\{{N \in \Sigma: \mu \left({N}\right) = 0}\right\}$, is denoted $\mathcal{N}_{\mu}$.

Definition in $\R^n$

A set $E \subseteq \R^n$ is called a null set if for any $\epsilon > 0$ there exists a countable collection $J_i := \left(({\mathbf{a}_i \,.\,.\, \mathbf{b}_i}\right))$, $i \in \N$ of open $n$-rectangles such that:

$\displaystyle E \subseteq \bigcup_{i = 1}^\infty J_i$

and

$\displaystyle \sum_{i = 1}^\infty \operatorname{vol} \left({J_i}\right) \leq \epsilon$.

Here, $\operatorname{vol} \left({J_i}\right)$ denotes the volume of the open rectangle $J_i$, which is the product of the lengths of its sides.

Said another way, a null set is a set that can be covered by a countable collection of open $n$-rectangles having total volume as small as we wish.

On Equivalence of Definitions of Null Set in Euclidean Space, it is shown that this definition is compatible with that for general measure spaces.

Note

Not to be confused with the empty set.

Sources

• René L. Schilling: Measures, Integrals and Martingales (2005)... (previous)... (next) $\S 4$: Problem $10$