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.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Someone told me today that if I can show $\Vert A_n-B_n\Vert_3\to 0$ as $n\to \infty$, then claiming $A=B$ as $n\to \infty$ (where $A$ and $B$ are the respective limits of $A_n$ and $B_n$) is a weaker claim than if I were to show that $\Vert A_n-B_n\Vert_2\to 0$ (which in turn is weaker than if I were to show $\Vert A_n-B_n\Vert_1\to 0$). Why is this so?

share|cite|improve this question
That would depend on what $A_n,B_n$ would be. – tomasz Oct 29 '12 at 2:03
@tomasz I don't follow... if need be, ignore $A_n$ and $B_n$ and just consider $\Vert A-B\Vert_3$ (and similarly, 2 and 1) – fkmasö Oct 29 '12 at 2:21
Presumably, you are getting at this:… – JavaMan Oct 29 '12 at 17:35

The answer pretty much depends on the measure space $ (X,\Sigma,\mu) $ that you are working with. There is the following well-known result (Rudin poses it as an exercise in his Real and Complex Analysis):

Let $ 1 \leq p < q \leq \infty $. Then

  1. $ {L^{q}}(X,\Sigma,\mu) \subseteq {L^{p}}(X,\Sigma,\mu) $ iff $ X $ does not have measurable subsets of arbitrarily large (but non-infinite) measure.

  2. $ {L^{q}}(X,\Sigma,\mu) \supseteq {L^{p}}(X,\Sigma,\mu) $ iff $ X $ does not have measurable subsets of arbitrarily small (but non-zero) measure.

If $ \mu(X) < \infty $, in which case (1) holds, we even have the following relationship between the norms $ \| \cdot \|_{p} $ and $ \| \cdot \|_{q} $: $$ \forall \, \text{measurable functions $ f $ on $ X $}: \quad \| f \|_{p} \leq [\mu(X)]^{\frac{1}{p} - \frac{1}{q}} \cdot \| f \|_{q}. $$ This yields the following sequence of implications: $$ \lim_{n \rightarrow \infty} \| f_{n} \|_{3} = 0 \quad \Longrightarrow \quad \lim_{n \rightarrow \infty} \| f_{n} \|_{2} = 0 \quad \Longrightarrow \quad \lim_{n \rightarrow \infty} \| f_{n} \|_{1} = 0. $$

I do not know what the conditions on $ (X,\Sigma,\mu) $ have to be in order to obtain the reverse of the above sequence of implications, namely, $$ \lim_{n \rightarrow \infty} \| f_{n} \|_{1} = 0 \quad \Longrightarrow \quad \lim_{n \rightarrow \infty} \| f_{n} \|_{2} = 0 \quad \Longrightarrow \quad \lim_{n \rightarrow \infty} \| f_{n} \|_{3} = 0. $$ This is the case that you are interested in. There may be results in interpolation theory (similar to the Riesz-Thorin Interpolation Theorem) that would give you precisely what you need.

share|cite|improve this answer

Your Answer


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.