Sequence such that every subsequence can have a different real limit [duplicate]
up vote
8
down vote
favorite
This question already has an answer here:
Give an example of a sequence of real numbers with subsequences converging to every real number
3 answers
I would like to find a sequence of real numbers $(a_n)_{ninmathbb{N}}$ with this property: for any $Linmathbb{R}$ there is a subsequence $a_{k_n}$ such that $$lim_{ntoinfty} a_{k_n} = L$$ Does such a sequence exist?
sequences-and-series limits analysis
marked as duplicate by Martin R, Community♦ Dec 9 at 14:29
This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.
add a comment |
up vote
8
down vote
favorite
This question already has an answer here:
Give an example of a sequence of real numbers with subsequences converging to every real number
3 answers
I would like to find a sequence of real numbers $(a_n)_{ninmathbb{N}}$ with this property: for any $Linmathbb{R}$ there is a subsequence $a_{k_n}$ such that $$lim_{ntoinfty} a_{k_n} = L$$ Does such a sequence exist?
sequences-and-series limits analysis
marked as duplicate by Martin R, Community♦ Dec 9 at 14:29
This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.
Otherwise said, the set ${ a_n | ninmathbb N}$ should be dense in $mathbb R$.
– Giuseppe Negro
Dec 9 at 11:49
add a comment |
up vote
8
down vote
favorite
up vote
8
down vote
favorite
This question already has an answer here:
Give an example of a sequence of real numbers with subsequences converging to every real number
3 answers
I would like to find a sequence of real numbers $(a_n)_{ninmathbb{N}}$ with this property: for any $Linmathbb{R}$ there is a subsequence $a_{k_n}$ such that $$lim_{ntoinfty} a_{k_n} = L$$ Does such a sequence exist?
sequences-and-series limits analysis
This question already has an answer here:
Give an example of a sequence of real numbers with subsequences converging to every real number
3 answers
I would like to find a sequence of real numbers $(a_n)_{ninmathbb{N}}$ with this property: for any $Linmathbb{R}$ there is a subsequence $a_{k_n}$ such that $$lim_{ntoinfty} a_{k_n} = L$$ Does such a sequence exist?
This question already has an answer here:
Give an example of a sequence of real numbers with subsequences converging to every real number
3 answers
sequences-and-series limits analysis
sequences-and-series limits analysis
edited Dec 9 at 11:46
Especially Lime
21.3k22656
21.3k22656
asked Dec 9 at 11:31
Riccardo Cazzin
1905
1905
marked as duplicate by Martin R, Community♦ Dec 9 at 14:29
This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.
marked as duplicate by Martin R, Community♦ Dec 9 at 14:29
This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.
Otherwise said, the set ${ a_n | ninmathbb N}$ should be dense in $mathbb R$.
– Giuseppe Negro
Dec 9 at 11:49
add a comment |
Otherwise said, the set ${ a_n | ninmathbb N}$ should be dense in $mathbb R$.
– Giuseppe Negro
Dec 9 at 11:49
Otherwise said, the set ${ a_n | ninmathbb N}$ should be dense in $mathbb R$.
– Giuseppe Negro
Dec 9 at 11:49
Otherwise said, the set ${ a_n | ninmathbb N}$ should be dense in $mathbb R$.
– Giuseppe Negro
Dec 9 at 11:49
add a comment |
4 Answers
4
active
oldest
votes
up vote
9
down vote
accepted
Just arrange the set of rational numbers in a sequence ${a_n}$. Given any real number $L$ and any positive integer $n$ there are infinitely many rationals in $(L-frac 1 n, L+frac 1 n)$. Pick $a_{n_1}$ in this interval with $n=1$. Then consider the case $n=2$. You can surely find $n_2 >n_1$ such that $a_{n_2} in (L-frac 1 2, L+frac 1 2)$. Use induction to construct a subsequence $a_{n_k}$ such that $|a_{n_k}-L| <frac 1 k$ for all $k$. Then $a_{n_k} to L$.
Could you please explain better how it would work? Thanks
– gimusi
Dec 9 at 11:40
That's nice, we are then using that rationals are countable. Thanks for the clarification. Regards
– gimusi
Dec 9 at 12:00
Most welcome, @gimusi !
– Kavi Rama Murthy
Dec 9 at 12:04
add a comment |
up vote
4
down vote
Take the sequence that sweep the interval $[-1,1]$ by $1/2$ steps, then the interval $[-2,2]$ by $1/2^2$ steps, then the interval $[-n,n]$ by $1/2^n$ steps... and so on.
You’ll be able to prove that every real is a limit point of that sequence.
That's less intuitive at first!
– gimusi
Dec 9 at 12:01
add a comment |
up vote
3
down vote
Rephrasing:
Consider $a_n$, $n=1,2,3,3,....,$ the sequence of rational numbers. Recall that $mathbb{Q}$ is countable, hence can be written as a sequence $a_n$, $nin mathbb{N}$.
Let $L in mathbb{R}$.
Since $mathbb{Q}$ is dense in $mathbb{R}$, we can construct a subsequence $a_{n_k}$ that converges to $L$.
add a comment |
up vote
1
down vote
Consider a general topological space $(X, mathscr{T})$ and a sequence of points $x in X^{mathbb{N}}$. One says that point $t in X$ is adherent to the sequence $x$ (some authors use the terminology 'cluster-point', but I don't fancy it so much) if:
$$(forall V, n)(V in mathscr{V}_{mathscr{T}}(x) wedge n in mathbb{N} implies (exists m)(m geqslant n wedge x_m in V ))$$
where $mathscr{V}_{mathscr{T}}(x)$ symbolizes the filter of neighbourhoods of point $x$ induced by the topology $mathscr{T}$. In a more descriptive fashion, $t$ is adherent to sequence $x$ if any neighbourhood of $t$ contains terms of arbitrarily high rank from the sequence $x$. If the filter of neighbourhoods of $t$ admits a countable base, then $t$ can be expressed as the limit of a subsequence of $x$. Therefore, in a space satisfying the First Axiom of Countability (i.e. all points have a countable base of neighbourhoods), the points adherent to a given sequence $x$ can be equivalently characterised as limits of subsequences of $x$.
Now, if the space $X$ is non-empty and separable, let us fix a certain dense subset $T subseteq X$. As $X$ is non-empty, so must $T$ be. It is not difficult to show that a non-empty countable set $M$ admits a surjection $sigma: mathbb{N} rightarrow M$ such that for each $t in M$ the fibre $sigma^{-1}({t})$ be infinite.
Consider such a surjection $sigma: mathbb{N} rightarrow T$ and define the sequence $t=(sigma(n))_{n in mathbb{N}}$ (which is actually the graphic of map $sigma$). The condition on the cardinality of the fibers ensures that any element $x in T$ is the limit of a (constant) subsequence of $t$. If we furthermore assume that the space $(X, mathscr{T})$ is $T_1$, then any $x in X setminus T$ will be an accumulation point of $T$ and thus necessarily adherent to sequence $t$.
To conclude, given a topological space $(X, mathscr{T})$ that is non-empty, separable, first countable and $T_1$, one can always find a sequence $t$ of points within the space such that each element $x in X$ be expressible as the limit of a subsequence of $t$. This applies in particular to the topological space $(mathbb{R}, mathscr{O})$, where $mathscr{O}$ denotes the (usual) order topology.
New contributor
add a comment |
4 Answers
4
active
oldest
votes
4 Answers
4
active
oldest
votes
active
oldest
votes
active
oldest
votes
up vote
9
down vote
accepted
Just arrange the set of rational numbers in a sequence ${a_n}$. Given any real number $L$ and any positive integer $n$ there are infinitely many rationals in $(L-frac 1 n, L+frac 1 n)$. Pick $a_{n_1}$ in this interval with $n=1$. Then consider the case $n=2$. You can surely find $n_2 >n_1$ such that $a_{n_2} in (L-frac 1 2, L+frac 1 2)$. Use induction to construct a subsequence $a_{n_k}$ such that $|a_{n_k}-L| <frac 1 k$ for all $k$. Then $a_{n_k} to L$.
Could you please explain better how it would work? Thanks
– gimusi
Dec 9 at 11:40
That's nice, we are then using that rationals are countable. Thanks for the clarification. Regards
– gimusi
Dec 9 at 12:00
Most welcome, @gimusi !
– Kavi Rama Murthy
Dec 9 at 12:04
add a comment |
up vote
9
down vote
accepted
Just arrange the set of rational numbers in a sequence ${a_n}$. Given any real number $L$ and any positive integer $n$ there are infinitely many rationals in $(L-frac 1 n, L+frac 1 n)$. Pick $a_{n_1}$ in this interval with $n=1$. Then consider the case $n=2$. You can surely find $n_2 >n_1$ such that $a_{n_2} in (L-frac 1 2, L+frac 1 2)$. Use induction to construct a subsequence $a_{n_k}$ such that $|a_{n_k}-L| <frac 1 k$ for all $k$. Then $a_{n_k} to L$.
Could you please explain better how it would work? Thanks
– gimusi
Dec 9 at 11:40
That's nice, we are then using that rationals are countable. Thanks for the clarification. Regards
– gimusi
Dec 9 at 12:00
Most welcome, @gimusi !
– Kavi Rama Murthy
Dec 9 at 12:04
add a comment |
up vote
9
down vote
accepted
up vote
9
down vote
accepted
Just arrange the set of rational numbers in a sequence ${a_n}$. Given any real number $L$ and any positive integer $n$ there are infinitely many rationals in $(L-frac 1 n, L+frac 1 n)$. Pick $a_{n_1}$ in this interval with $n=1$. Then consider the case $n=2$. You can surely find $n_2 >n_1$ such that $a_{n_2} in (L-frac 1 2, L+frac 1 2)$. Use induction to construct a subsequence $a_{n_k}$ such that $|a_{n_k}-L| <frac 1 k$ for all $k$. Then $a_{n_k} to L$.
Just arrange the set of rational numbers in a sequence ${a_n}$. Given any real number $L$ and any positive integer $n$ there are infinitely many rationals in $(L-frac 1 n, L+frac 1 n)$. Pick $a_{n_1}$ in this interval with $n=1$. Then consider the case $n=2$. You can surely find $n_2 >n_1$ such that $a_{n_2} in (L-frac 1 2, L+frac 1 2)$. Use induction to construct a subsequence $a_{n_k}$ such that $|a_{n_k}-L| <frac 1 k$ for all $k$. Then $a_{n_k} to L$.
edited Dec 9 at 11:45
answered Dec 9 at 11:37
Kavi Rama Murthy
46.6k31854
46.6k31854
Could you please explain better how it would work? Thanks
– gimusi
Dec 9 at 11:40
That's nice, we are then using that rationals are countable. Thanks for the clarification. Regards
– gimusi
Dec 9 at 12:00
Most welcome, @gimusi !
– Kavi Rama Murthy
Dec 9 at 12:04
add a comment |
Could you please explain better how it would work? Thanks
– gimusi
Dec 9 at 11:40
That's nice, we are then using that rationals are countable. Thanks for the clarification. Regards
– gimusi
Dec 9 at 12:00
Most welcome, @gimusi !
– Kavi Rama Murthy
Dec 9 at 12:04
Could you please explain better how it would work? Thanks
– gimusi
Dec 9 at 11:40
Could you please explain better how it would work? Thanks
– gimusi
Dec 9 at 11:40
That's nice, we are then using that rationals are countable. Thanks for the clarification. Regards
– gimusi
Dec 9 at 12:00
That's nice, we are then using that rationals are countable. Thanks for the clarification. Regards
– gimusi
Dec 9 at 12:00
Most welcome, @gimusi !
– Kavi Rama Murthy
Dec 9 at 12:04
Most welcome, @gimusi !
– Kavi Rama Murthy
Dec 9 at 12:04
add a comment |
up vote
4
down vote
Take the sequence that sweep the interval $[-1,1]$ by $1/2$ steps, then the interval $[-2,2]$ by $1/2^2$ steps, then the interval $[-n,n]$ by $1/2^n$ steps... and so on.
You’ll be able to prove that every real is a limit point of that sequence.
That's less intuitive at first!
– gimusi
Dec 9 at 12:01
add a comment |
up vote
4
down vote
Take the sequence that sweep the interval $[-1,1]$ by $1/2$ steps, then the interval $[-2,2]$ by $1/2^2$ steps, then the interval $[-n,n]$ by $1/2^n$ steps... and so on.
You’ll be able to prove that every real is a limit point of that sequence.
That's less intuitive at first!
– gimusi
Dec 9 at 12:01
add a comment |
up vote
4
down vote
up vote
4
down vote
Take the sequence that sweep the interval $[-1,1]$ by $1/2$ steps, then the interval $[-2,2]$ by $1/2^2$ steps, then the interval $[-n,n]$ by $1/2^n$ steps... and so on.
You’ll be able to prove that every real is a limit point of that sequence.
Take the sequence that sweep the interval $[-1,1]$ by $1/2$ steps, then the interval $[-2,2]$ by $1/2^2$ steps, then the interval $[-n,n]$ by $1/2^n$ steps... and so on.
You’ll be able to prove that every real is a limit point of that sequence.
answered Dec 9 at 11:42
mathcounterexamples.net
23.9k21753
23.9k21753
That's less intuitive at first!
– gimusi
Dec 9 at 12:01
add a comment |
That's less intuitive at first!
– gimusi
Dec 9 at 12:01
That's less intuitive at first!
– gimusi
Dec 9 at 12:01
That's less intuitive at first!
– gimusi
Dec 9 at 12:01
add a comment |
up vote
3
down vote
Rephrasing:
Consider $a_n$, $n=1,2,3,3,....,$ the sequence of rational numbers. Recall that $mathbb{Q}$ is countable, hence can be written as a sequence $a_n$, $nin mathbb{N}$.
Let $L in mathbb{R}$.
Since $mathbb{Q}$ is dense in $mathbb{R}$, we can construct a subsequence $a_{n_k}$ that converges to $L$.
add a comment |
up vote
3
down vote
Rephrasing:
Consider $a_n$, $n=1,2,3,3,....,$ the sequence of rational numbers. Recall that $mathbb{Q}$ is countable, hence can be written as a sequence $a_n$, $nin mathbb{N}$.
Let $L in mathbb{R}$.
Since $mathbb{Q}$ is dense in $mathbb{R}$, we can construct a subsequence $a_{n_k}$ that converges to $L$.
add a comment |
up vote
3
down vote
up vote
3
down vote
Rephrasing:
Consider $a_n$, $n=1,2,3,3,....,$ the sequence of rational numbers. Recall that $mathbb{Q}$ is countable, hence can be written as a sequence $a_n$, $nin mathbb{N}$.
Let $L in mathbb{R}$.
Since $mathbb{Q}$ is dense in $mathbb{R}$, we can construct a subsequence $a_{n_k}$ that converges to $L$.
Rephrasing:
Consider $a_n$, $n=1,2,3,3,....,$ the sequence of rational numbers. Recall that $mathbb{Q}$ is countable, hence can be written as a sequence $a_n$, $nin mathbb{N}$.
Let $L in mathbb{R}$.
Since $mathbb{Q}$ is dense in $mathbb{R}$, we can construct a subsequence $a_{n_k}$ that converges to $L$.
answered Dec 9 at 11:51
Peter Szilas
10.5k2720
10.5k2720
add a comment |
add a comment |
up vote
1
down vote
Consider a general topological space $(X, mathscr{T})$ and a sequence of points $x in X^{mathbb{N}}$. One says that point $t in X$ is adherent to the sequence $x$ (some authors use the terminology 'cluster-point', but I don't fancy it so much) if:
$$(forall V, n)(V in mathscr{V}_{mathscr{T}}(x) wedge n in mathbb{N} implies (exists m)(m geqslant n wedge x_m in V ))$$
where $mathscr{V}_{mathscr{T}}(x)$ symbolizes the filter of neighbourhoods of point $x$ induced by the topology $mathscr{T}$. In a more descriptive fashion, $t$ is adherent to sequence $x$ if any neighbourhood of $t$ contains terms of arbitrarily high rank from the sequence $x$. If the filter of neighbourhoods of $t$ admits a countable base, then $t$ can be expressed as the limit of a subsequence of $x$. Therefore, in a space satisfying the First Axiom of Countability (i.e. all points have a countable base of neighbourhoods), the points adherent to a given sequence $x$ can be equivalently characterised as limits of subsequences of $x$.
Now, if the space $X$ is non-empty and separable, let us fix a certain dense subset $T subseteq X$. As $X$ is non-empty, so must $T$ be. It is not difficult to show that a non-empty countable set $M$ admits a surjection $sigma: mathbb{N} rightarrow M$ such that for each $t in M$ the fibre $sigma^{-1}({t})$ be infinite.
Consider such a surjection $sigma: mathbb{N} rightarrow T$ and define the sequence $t=(sigma(n))_{n in mathbb{N}}$ (which is actually the graphic of map $sigma$). The condition on the cardinality of the fibers ensures that any element $x in T$ is the limit of a (constant) subsequence of $t$. If we furthermore assume that the space $(X, mathscr{T})$ is $T_1$, then any $x in X setminus T$ will be an accumulation point of $T$ and thus necessarily adherent to sequence $t$.
To conclude, given a topological space $(X, mathscr{T})$ that is non-empty, separable, first countable and $T_1$, one can always find a sequence $t$ of points within the space such that each element $x in X$ be expressible as the limit of a subsequence of $t$. This applies in particular to the topological space $(mathbb{R}, mathscr{O})$, where $mathscr{O}$ denotes the (usual) order topology.
New contributor
add a comment |
up vote
1
down vote
Consider a general topological space $(X, mathscr{T})$ and a sequence of points $x in X^{mathbb{N}}$. One says that point $t in X$ is adherent to the sequence $x$ (some authors use the terminology 'cluster-point', but I don't fancy it so much) if:
$$(forall V, n)(V in mathscr{V}_{mathscr{T}}(x) wedge n in mathbb{N} implies (exists m)(m geqslant n wedge x_m in V ))$$
where $mathscr{V}_{mathscr{T}}(x)$ symbolizes the filter of neighbourhoods of point $x$ induced by the topology $mathscr{T}$. In a more descriptive fashion, $t$ is adherent to sequence $x$ if any neighbourhood of $t$ contains terms of arbitrarily high rank from the sequence $x$. If the filter of neighbourhoods of $t$ admits a countable base, then $t$ can be expressed as the limit of a subsequence of $x$. Therefore, in a space satisfying the First Axiom of Countability (i.e. all points have a countable base of neighbourhoods), the points adherent to a given sequence $x$ can be equivalently characterised as limits of subsequences of $x$.
Now, if the space $X$ is non-empty and separable, let us fix a certain dense subset $T subseteq X$. As $X$ is non-empty, so must $T$ be. It is not difficult to show that a non-empty countable set $M$ admits a surjection $sigma: mathbb{N} rightarrow M$ such that for each $t in M$ the fibre $sigma^{-1}({t})$ be infinite.
Consider such a surjection $sigma: mathbb{N} rightarrow T$ and define the sequence $t=(sigma(n))_{n in mathbb{N}}$ (which is actually the graphic of map $sigma$). The condition on the cardinality of the fibers ensures that any element $x in T$ is the limit of a (constant) subsequence of $t$. If we furthermore assume that the space $(X, mathscr{T})$ is $T_1$, then any $x in X setminus T$ will be an accumulation point of $T$ and thus necessarily adherent to sequence $t$.
To conclude, given a topological space $(X, mathscr{T})$ that is non-empty, separable, first countable and $T_1$, one can always find a sequence $t$ of points within the space such that each element $x in X$ be expressible as the limit of a subsequence of $t$. This applies in particular to the topological space $(mathbb{R}, mathscr{O})$, where $mathscr{O}$ denotes the (usual) order topology.
New contributor
add a comment |
up vote
1
down vote
up vote
1
down vote
Consider a general topological space $(X, mathscr{T})$ and a sequence of points $x in X^{mathbb{N}}$. One says that point $t in X$ is adherent to the sequence $x$ (some authors use the terminology 'cluster-point', but I don't fancy it so much) if:
$$(forall V, n)(V in mathscr{V}_{mathscr{T}}(x) wedge n in mathbb{N} implies (exists m)(m geqslant n wedge x_m in V ))$$
where $mathscr{V}_{mathscr{T}}(x)$ symbolizes the filter of neighbourhoods of point $x$ induced by the topology $mathscr{T}$. In a more descriptive fashion, $t$ is adherent to sequence $x$ if any neighbourhood of $t$ contains terms of arbitrarily high rank from the sequence $x$. If the filter of neighbourhoods of $t$ admits a countable base, then $t$ can be expressed as the limit of a subsequence of $x$. Therefore, in a space satisfying the First Axiom of Countability (i.e. all points have a countable base of neighbourhoods), the points adherent to a given sequence $x$ can be equivalently characterised as limits of subsequences of $x$.
Now, if the space $X$ is non-empty and separable, let us fix a certain dense subset $T subseteq X$. As $X$ is non-empty, so must $T$ be. It is not difficult to show that a non-empty countable set $M$ admits a surjection $sigma: mathbb{N} rightarrow M$ such that for each $t in M$ the fibre $sigma^{-1}({t})$ be infinite.
Consider such a surjection $sigma: mathbb{N} rightarrow T$ and define the sequence $t=(sigma(n))_{n in mathbb{N}}$ (which is actually the graphic of map $sigma$). The condition on the cardinality of the fibers ensures that any element $x in T$ is the limit of a (constant) subsequence of $t$. If we furthermore assume that the space $(X, mathscr{T})$ is $T_1$, then any $x in X setminus T$ will be an accumulation point of $T$ and thus necessarily adherent to sequence $t$.
To conclude, given a topological space $(X, mathscr{T})$ that is non-empty, separable, first countable and $T_1$, one can always find a sequence $t$ of points within the space such that each element $x in X$ be expressible as the limit of a subsequence of $t$. This applies in particular to the topological space $(mathbb{R}, mathscr{O})$, where $mathscr{O}$ denotes the (usual) order topology.
New contributor
Consider a general topological space $(X, mathscr{T})$ and a sequence of points $x in X^{mathbb{N}}$. One says that point $t in X$ is adherent to the sequence $x$ (some authors use the terminology 'cluster-point', but I don't fancy it so much) if:
$$(forall V, n)(V in mathscr{V}_{mathscr{T}}(x) wedge n in mathbb{N} implies (exists m)(m geqslant n wedge x_m in V ))$$
where $mathscr{V}_{mathscr{T}}(x)$ symbolizes the filter of neighbourhoods of point $x$ induced by the topology $mathscr{T}$. In a more descriptive fashion, $t$ is adherent to sequence $x$ if any neighbourhood of $t$ contains terms of arbitrarily high rank from the sequence $x$. If the filter of neighbourhoods of $t$ admits a countable base, then $t$ can be expressed as the limit of a subsequence of $x$. Therefore, in a space satisfying the First Axiom of Countability (i.e. all points have a countable base of neighbourhoods), the points adherent to a given sequence $x$ can be equivalently characterised as limits of subsequences of $x$.
Now, if the space $X$ is non-empty and separable, let us fix a certain dense subset $T subseteq X$. As $X$ is non-empty, so must $T$ be. It is not difficult to show that a non-empty countable set $M$ admits a surjection $sigma: mathbb{N} rightarrow M$ such that for each $t in M$ the fibre $sigma^{-1}({t})$ be infinite.
Consider such a surjection $sigma: mathbb{N} rightarrow T$ and define the sequence $t=(sigma(n))_{n in mathbb{N}}$ (which is actually the graphic of map $sigma$). The condition on the cardinality of the fibers ensures that any element $x in T$ is the limit of a (constant) subsequence of $t$. If we furthermore assume that the space $(X, mathscr{T})$ is $T_1$, then any $x in X setminus T$ will be an accumulation point of $T$ and thus necessarily adherent to sequence $t$.
To conclude, given a topological space $(X, mathscr{T})$ that is non-empty, separable, first countable and $T_1$, one can always find a sequence $t$ of points within the space such that each element $x in X$ be expressible as the limit of a subsequence of $t$. This applies in particular to the topological space $(mathbb{R}, mathscr{O})$, where $mathscr{O}$ denotes the (usual) order topology.
New contributor
New contributor
answered Dec 9 at 14:09
ΑΘΩ
2463
2463
New contributor
New contributor
add a comment |
add a comment |
Otherwise said, the set ${ a_n | ninmathbb N}$ should be dense in $mathbb R$.
– Giuseppe Negro
Dec 9 at 11:49