Inclusion of limit point in Collection set

98 Views Asked by At

Let $F$ be collection of set in $R^n$ and let $S=\cup_{A\in F}A$ and $T=\cap_{A\in F}A$.
Then prove or disprove following facts.
1) If x is limit point of T,then x is limit point of each of $A\in F$
2) If x is limit point of S,then x is limit point of atleast one of $A\in F$.

My Attempt: 1) x is limit point of $\cap_{A\in F}A$ then for $\forall \epsilon>0 \exists y\in $$\cap_{A\in F}A$ i.e $y \in $every$ A$ Hence x is limit point of every A.
2)x is limit point of $\cup_{A\in F}A$ then for $\forall \epsilon>0 \exists y\in $ A for some $A \in F$ Hence done.
Actually both statement looking like trivial.
I just wanted to confirm are they true.Or there exists some counterexample.
ANy Help will be appreciated

1

There are 1 best solutions below

0
On BEST ANSWER

If the statements are looking trivial, then you are not looking carefully!

What does it mean to be a limit point of a set? $x$ is a limit point of a set $B$ if for all $\epsilon > 0$ there exists $y \in B$ such that $d(y,x) < \epsilon$. (Sometimes, $x \neq y$ is also insisted).

The intersection is done correctly : if $x$ is a limit point of $\cap_{A \in F} A$, then for all $\epsilon > 0$ there is $y \in \cap_{A \in F} A$, such that $d(y,x) < \epsilon$. Now, note that $y \in A$ for all $A$, so if $\epsilon > 0$ then this choice of $y$ works to show that $y \in A$ and $d(y,x) < \epsilon$. Consequently, $x$ is a limit point of each $A$.


However, the union is not done correctly, and here's why : fix $\epsilon_1 > 0$. What we know, is that $x$ is a limit point of $\cup_{A \in F} A$, so there is some $y \in A_1 (\in F)$ such that $d(y,x) < \epsilon$. Note that for some other $\epsilon_2$, there may be some other $A_2$, and for some other $\epsilon_3$ some other $A_3$ : in short, it is possible that for all $A \in F$, all $\epsilon$ below some point stop working out after some time. That is, it is possible that for all $A \in F$ , there exists $\epsilon_A > 0$ such that for all $y \in A$, we have $d(y,x) > \epsilon_A$. So $x$ would not be a limit point of any of the $A$ but be a limit point of their union.

The best example of this, is a convergent sequence broken into parts : for example, let $A_n = \{\frac 1n\}$ be singleton sets with the element $\frac 1n$. Then, if you take the union of $A_n$, you get the set $\{1,\frac 12,\frac 13,...\}$ which has a limit point $0$. But $0$ is not a limit point of any $A_n$.


However, you can check that if $F$ is a finite set of sets, then actually this fact is true. See if you can figure out why.