I just could not convince myself about what $\chi_{\cup_{A_n}}$ is worth, if $(A_n)$ is a sequence of pairwise disjoint sets? Is it equal to the series $\sum_{n=1}^{\infty}\chi_{A_n}$? We need to investigate the convergence of the series but how? Many thanks.


Clearly, $\chi_{\bigcup_{n=1}^{\infty}A_n}(x)$ can assume only two values (it's a characteristic function after all): $1$ or $0$ according as $x\in \bigcup_{n=1}^{\infty}A_n$ or $x\notin\bigcup_{n=1}^{\infty}A_n$.

If $x\in \bigcup_{n=1}^{\infty}A_n$, then $x\in A_m$ for exactly one $m\in\mathbb N$ (given that the union is disjoint), so that $\sum_{n=1}^{\infty}\chi_{A_n}(x)=\chi_{A_m}(x)=1$ (and the series converges because its members vanish for $n>m$). On the other hand, if $x\notin \bigcup_{n=1}^{\infty}A_n$, then $x\in A_m$ for no $m\in\mathbb N$, which implies that $\sum_{n=1}^{\infty}\chi_{A_n}(x)=0$ (and the series converges, since all of its members vanish).

Conclusion: $\chi_{\bigcup_{n=1}^{\infty}A_n}=\sum_{n=1}^{\infty}\chi_{A_n}$, indeed.

  • $\begingroup$ Thanks a lot triple_sec. I had this idea and thanks for your detailed solution. I posted this as regards to some spectral measure in Rudin...I think I must have misunderstood Rudin's argument as the two contexts are little different. Thanks a lot again. $\endgroup$
    – Math
    May 25 '15 at 8:33
  • $\begingroup$ @Math You're most welcome! $\endgroup$
    – triple_sec
    May 25 '15 at 8:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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