MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

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

Give an example satisfying the following conditions: give out a sequence of random variables defined on a probability space, and a sub sigma algebra: the sequence converges almost surely to a limit and it is also uniformly integrable, but the sequence of the conditional expectation of the random variable sequence on the sub sigma algebra does not converges almost surely to the limit, which is the conditional expectation of the limit of the random variable sequence on the sub sigma algebra.

share|cite|improve this question

closed as too localized by Nate Eldredge, Andrés E. Caicedo, George Lowther, Zev Chonoles, Mark Sapir Feb 23 '11 at 4:07

This question is unlikely to help any future visitors; it is only relevant to a small geographic area, a specific moment in time, or an extraordinarily narrow situation that is not generally applicable to the worldwide audience of the internet. For help making this question more broadly applicable, visit the help center.If this question can be reworded to fit the rules in the help center, please edit the question.

This isn't really a research level question (see the faq: Actually, it's not a question, rather, its a command. – George Lowther Feb 23 '11 at 3:30
math.stackexchange is probably a more suitable site although, if you post it there, it should still be written as a question. – George Lowther Feb 23 '11 at 3:31

Consider independent random variables $X_n, Y_n, n\in\mathbb N$, such that $\mathbb P(X_n = 1) = 1-\mathbb P(X_n = 0) = 1/n$, $\mathbb P(Y_n = n) = 1-\mathbb P(Y_n = 0) = 1/n$. Set $Z_n = X_nY_n$. Then, $Z_n$ is uniformly integrable and $Z_n\to 0$ as $n\to\infty$ almost surely, by Borel-Cantelli Lemma. However, set ${\cal A} = \sigma(X_n:n\in\mathbb N)$. Then $\mathbb E(Z_n\mid{\cal A}) = X_n\mathbb E(Y_n\mid {\cal A}) = X_n$, which converges to 0 in $L^1$ but not almost surely.

share|cite|improve this answer

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