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.

Let $\Omega$ be a set, $\mathcal B$ is a semi-algebra that contains $\Omega$, and let $\mu \colon \mathcal B\rightarrow [0,\infty]$ be a measure defined on $(\Omega,B)$. Now define the algebra $\mathcal A$ which contains all the finite (disjoint) unions of sets from $\mathcal B$: $$\mathcal A=\left\{E=E_1 \cup E_2 \cup\cdots\cup E_n: E_1,E_2,\cdots,E_n\in \mathcal B, E_i\cap E_j=\varnothing \right\}.$$ Now define $\nu\colon\mathcal A\rightarrow [0,\infty]$ as $\nu (E)=\sum_{k=1}^n \mu (E_k)$. I shall prove that $\nu$ is a measure on $\mathcal A$.

I'm having problems with showing the $\sigma$-additivity property. How do I show that if $Q_1,Q_2,\ldots\in\mathcal A$ and $\bigcup_{n=1}^\infty Q_n \in \mathcal A$ then $\nu\left(\bigcup_{n=1}^\infty Q_n\right)=\sum_n \nu(Q_n)$?

share|improve this question
1  
You assume that the sets $Q_n$ are pairwise disjoint. If $\bigcup_{n=1}^{+\infty}Q_n\in A$, write it as $E_1\cup\cdots\cup E_N$ where $E_i\in\mathcal A$ and are pairwise disjoint. Write $Q_n:=\bigcup_{j=1}^{N_n}E_{j,n}$. The sets $E_i\cap E_{j,n}$ are pairwise disjoint. –  Davide Giraudo Nov 27 '11 at 21:41
    
Please don't include things like "HW" in the subject line; that's what the [homework] tag is for. Thank you. –  Arturo Magidin Nov 27 '11 at 22:00
    
@Davide Giraudo: but what from here? I write that $\nu (\cup_n Q_n)=\sum_{k=1}^N \mu (E_k) = \sum_{k=1}^N \mu (\cup_{n=1}^\infty \cup_{j=1}^{N_n} E_k\cap E_{j,n} )$ but I cant continue further because I'm not sure that $\cup_{j=1}^{N_n} E_k\cap E_{j,n} \in B$ –  Max Nov 27 '11 at 22:22
    
@Max: Isn't a semialgebra necessarily closed under binary intersections? This is true for semirings, and a semialgebra is just a semiring that includes the whole set. –  Arturo Magidin Nov 27 '11 at 23:40
    
@Arturo Magidin: I'm sorry, but I dont know what are binary intersections. Can I prove the theorem without them (or without using the fact that its closed under these intersections)? –  Max Nov 27 '11 at 23:56
show 1 more comment

Your Answer

 
discard

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

Browse other questions tagged or ask your own question.