Convergence in Measure Implies Integrable

718 Views Asked by At

Let $f_n$ be a sequence of measurable functions which converge in measure to a function $f$. My first question is, is $f$ itself necessarily measurable?

Now suppose that $|f_n| \leq |g|$ for some integrable function $g$. How do we argue that $|f-f_n|$ is integrable? If $f_n$ converged pointwise to $f$, then we would also have $|f| \leq |g|$, so that $|f- f_n| \leq 2 |g|$, but we can't do that here. The textbook I'm using is Royden, by the way.

1

There are 1 best solutions below

3
On BEST ANSWER

Yes the limiting function is measurable. To prove this, show that convergence in measure implies convergence almost surely along a subsequence. Then use the fact (or reprove easily) that almost sure convergence implies measurability. As a hint, recall that you would need to show $\{x: \ f(x)\leq t\}$ is measurable for each $t$. If you focus on the set $A$ where $f_n(x)$ converges to $x$, then then you can write $\{x: \ f(x)\leq t\}$ as $\cap_n \{x: f_n(x)\leq t\}$ plus some measure 0 set.

To se that $|f-f_n|$ is integrable, use triangle inequality: $|f-f_n|\leq |f|+|f_n|\leq |f|+|g|\leq 2|g|$, where the r.h.s. is integrable. You are using the fact that if $|f|\leq |g|$, where $|g|$ is integrable, then so is $f(x)$ (assuming both are measurable of course).