Let $(X,m)$ be a metric measure space, $(f_n)_n$ a sequence in $L^\infty, f \in L^\infty$ s.t. $$ \int gf_n \ dm \rightarrow \int g f $$ for every $g : X \rightarrow \mathbb R$ continuous and bounded. I have seen in a book the claim that if $ \forall n, \|f_n\|_{L^\infty} \leq C$ then $$ \int hf_n \ dm \rightarrow \int h f ,$$ for every $h \in L^1$.

How can I prove this result?


This is true provided that $C_b \cap L^1$ is dense in $L^1$, which holds if $m$ is a finite measure, or if $X$ is locally compact and $m$ is Radon (I imagine many other conditions are possible). In this case, we can use a standard triangle inequality trick. Given any $h \in L^1$ and any $\epsilon > 0$ we can find $g \in C_b \cap L_1$ such that $\|g-h\|_{L^1} < \epsilon$. Then $$\begin{align*}\left| \int f_n h\,dm - \int f h\,dm \right| &= \left| \int (f_n h - f_n g + f_n g - f g + f g - f h)\,dm \right| \\ &\le \left| \int f_n (h-g)\,dm \right| + \left| \int (f_n - f) g\,dm \right| + \left| \int f (g-h)\,dm \right| \end{align*}$$ The first term is bounded by $\|f_n\|_{L^\infty} \|g-h\|_{L^1} < C \epsilon$. The second term goes to 0 by assumption. The third term is bounded by $\|f\|_{L^\infty} \|g-h\|_{L^1} < \|f\|_{L^\infty} \epsilon$. Hence $$\limsup_{n \to \infty} \left| \int f_n h\,dm - \int f h\,dm \right| \le C \epsilon + \|f\|_{L^\infty} \epsilon + \limsup_{n \to \infty} \left| \int (f_n - f) g\,dm \right| = C \epsilon + \|f\|_{L^\infty} \epsilon.$$ Since $\epsilon$ was arbitrary the limsup on the left must be 0. It is a nonnegative sequence, so it must actually converge to 0, which is what was desired.

I am not exactly sure what can happen if $C_b \cap L^1$ is not dense in $L^1$.

  • $\begingroup$ Thank you! However this does not work in spaces where continuous functions are not dense in $L^1$, but I guess they are too ugly anyway... $\endgroup$ – Brainstorming Jun 3 '15 at 12:47
  • $\begingroup$ @Brainstorming: Thanks for the correction. I missed that we were working on an arbitrary metric space - I saw an $m$ and assumed it was Lebesgue measure. $\endgroup$ – Nate Eldredge Jun 3 '15 at 13:50

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.