Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

For $i=1,2$, let $\Sigma_i$ be two $\sigma$-algebras of subsets of some sets $S_i$. Denote by $\Sigma_1\otimes\Sigma_2$ their product $\sigma$-algebra. It is well known that any bounded, $\Sigma_1\otimes\Sigma_2$-measurable function can be approximated in the $\sup$-norm by simple functions $\sum_{k=1}^{n}a_k\cdot 1_{A_k}$, where $A_k\in \Sigma_1\otimes\Sigma_2$. Let $A\in \Sigma_1\otimes\Sigma_2$. Is it possible to approximate the characteristic function $1_A$ by simple functions of the form $1_{E\times F}$, where $E\in\Sigma_1$ and $F\in\Sigma_2$, i.e. for any $\varepsilon>0$, does there exist $n\in\mathbb{N}$, real numbers $a_1,...,a_n\in\mathbb{R}$ and sets $E_k\times F_k\in \Sigma_1\times\Sigma_2$, $k=1,...,n$ such that $$\sup_{(x,y)\in S_1\times S_2}\vert1_A(x,y)-\sum_{k=1}^{n}a_k\cdot 1_{E_k\times F_k}(x,y)\vert <\varepsilon\ ?$$

share|improve this question
1  
No ... just consider $\Delta_{\mathbb R} := \{(x,x) \mid x \in \mathbb R\} \subseteq \mathbb R^2$. –  martini Apr 19 '12 at 12:19
add comment

1 Answer

up vote 1 down vote accepted

This approximation is true if and only if $A$ is a finite union of sets of the form $A_1\times A_2$, where $A_i\in\Sigma_i$, $i\in [2]$.

Indeed, the condition is trivially sufficient, to see it's necessary take $\varepsilon=\frac 12$. We can assume that $\{E_k\}_{k=1}^n, \{F_k\}_{k=1}^n$ are pairwise disjoint families of measurable sets and $a_k\geq 0$. Then if $(x,y)\in A$, we have $$\frac 12< \sum_{k=1}^na_k\chi_{E_k}(x)\chi_{F_k}(y)< \frac 32.$$ In particular $(x,y)\in\bigcup_{k=1}^nE_k\times F_k$ so $A\subset \bigcup_{k=1}^nE_k\times F_k$. The displayed equation shows that $a_k\geq \frac 12$ for each $k\in [n]$. If $(x,y)\in E_i\times F_i$ for some $i$, we should have $|\chi_A(x,y)-a_i|<\frac 12$ and this implies $(x,y)\in A$.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.