I got stuck on this problem, maybe someone can give me a hint
Let (X, M , m) be a measure space and let (f_n: X --> [0, oo]) be a
sequence of functions that converges to a function f. Suppose that lim (\Integral(over X) f_n dm) = \Integral(over X) f dm < oo. Then, for every set E in the sigma-algebra M, we have lim (\Integral(over E) f_n dm) = \Integral(over E) f dm. Show that this conclusion may fail if \Integral(over X) f dm = oo.
If lim (\Integral(over E) f_n dm) = \Integral(over E) f dm is not true, then it's not hard to show that there's a subsequence (f_n_k) such that lim (\Integral(over E) f_n_k dm) < \Integral(over E) f dm, but I don't see why this leads to a contadiction.
Thank you
2006-09-25
10:17:10
·
2 answers
·
asked by
Steiner
7
in
Science & Mathematics
➔ Mathematics