Non-convergence written with sets

  • Context: Graduate 
  • Thread starter Thread starter cuak2000
  • Start date Start date
  • Tags Tags
    Sets
Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
1 reply · 2K views
cuak2000
Messages
7
Reaction score
0
Hey, everyone. I'm trying to prove the following:

[tex]f_n[/tex] and [tex]f_n[/tex] are real-valued function in [tex]\Omega[/tex]

[tex]\{\omega: f_n(\omega) \nrightarrow f(\omega) \} = \\<br /> \bigcup^{\infty}_{k=1} \bigcap^{\infty}_{N=1} \bigcup^{\infty}_{n=1} <br /> \{ \omega : | f_n(\omega) - f(\omega) | \geq 1/k \}<br /> <br /> [/tex]

I am convinced by the proof I've made up, but it isn't formal, so I would appreciate if you could help me give it more formality.
Let's call the left side of the equality L and the right side R.
L can be written:

[tex]\exists k \in \mathbb{N} \quad \forall N \quad \exists n \geq N \quad | f_n(\omega) \nrightarrow f(\omega) | \geq 1/k[/tex]

On the other hand, the last part of R is
[tex]\bigcap^{\infty}_{N=1} \bigcup^{\infty}_{n=1} <br /> \{ \omega : | f_n(\omega) - f(\omega) | \geq 1/k \}[/tex]

which basically takes all the [tex]\omega[/tex] that [tex]\forall N[/tex] have
at least one [tex]n \geq N[/tex] that makes the absolute difference bigger than 1/k
If you take the union for all k, then you have the definition for being in L.
Thanks in advance,

cd
 
Last edited:
Physics news on Phys.org
There's something wrong with your R. You take an intersection over N=1 to infty, but the index N does not appear in the collection over which the intersection is taken. You probably meant

[tex]\bigcup^{\infty}_{k=1} \bigcap^{\infty}_{N=1} \bigcup^{\infty}_{n=N} <br /> \{ \omega : | f_n(\omega) - f(\omega) | \geq 1/k \}[/tex]

and then you are already done, since the condition for omega to belong to this one and to to LHS are the same, namely

[tex]\exists k \in \mathbb{N} \quad \forall N \quad \exists n \geq N \quad | f_n(\omega) \nrightarrow f(\omega) | \geq 1/k[/tex]