Convergence of $a_n$ given $a_{lfloor{x^n}rfloor}$ converges to $0$ [duplicate]
This question already has an answer here:
Convergence of a sequence with assumption that exponential subsequences converge?
1 answer
Sequence $a_n$ has a property that for every real $x > 1$, sequence $a_{lfloor{x^n}rfloor}$ converges to $0$. Does that mean that $a_n$ converges to $0$? I have tried to find a counterexample, but failed at it, so I think it might be true, but I don't know how to prove it.
real-analysis limits convergence
marked as duplicate by gimusi
StackExchange.ready(function() {
if (StackExchange.options.isMobile) return;
$('.dupe-hammer-message-hover:not(.hover-bound)').each(function() {
var $hover = $(this).addClass('hover-bound'),
$msg = $hover.siblings('.dupe-hammer-message');
$hover.hover(
function() {
$hover.showInfoMessage('', {
messageElement: $msg.clone().show(),
transient: false,
position: { my: 'bottom left', at: 'top center', offsetTop: -7 },
dismissable: false,
relativeToBody: true
});
},
function() {
StackExchange.helpers.removeMessages();
}
);
});
});
Nov 28 at 8:08
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 |
This question already has an answer here:
Convergence of a sequence with assumption that exponential subsequences converge?
1 answer
Sequence $a_n$ has a property that for every real $x > 1$, sequence $a_{lfloor{x^n}rfloor}$ converges to $0$. Does that mean that $a_n$ converges to $0$? I have tried to find a counterexample, but failed at it, so I think it might be true, but I don't know how to prove it.
real-analysis limits convergence
marked as duplicate by gimusi
StackExchange.ready(function() {
if (StackExchange.options.isMobile) return;
$('.dupe-hammer-message-hover:not(.hover-bound)').each(function() {
var $hover = $(this).addClass('hover-bound'),
$msg = $hover.siblings('.dupe-hammer-message');
$hover.hover(
function() {
$hover.showInfoMessage('', {
messageElement: $msg.clone().show(),
transient: false,
position: { my: 'bottom left', at: 'top center', offsetTop: -7 },
dismissable: false,
relativeToBody: true
});
},
function() {
StackExchange.helpers.removeMessages();
}
);
});
});
Nov 28 at 8:08
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.
$a_n$ converges to 0 is equivalent to the following proposition : For every sequence $u_n$, there exists a real $x > 1$ such that ${u_n}_{n in mathbb{N}} cap {lfloor{x^n}rfloor}_{n in mathbb{N}}$ is infinite. I'm not sure how to procede from here.
– Rchn
Nov 27 at 13:26
For every *strictly increasing sequence $u_n$.
– Rchn
Nov 27 at 14:34
add a comment |
This question already has an answer here:
Convergence of a sequence with assumption that exponential subsequences converge?
1 answer
Sequence $a_n$ has a property that for every real $x > 1$, sequence $a_{lfloor{x^n}rfloor}$ converges to $0$. Does that mean that $a_n$ converges to $0$? I have tried to find a counterexample, but failed at it, so I think it might be true, but I don't know how to prove it.
real-analysis limits convergence
This question already has an answer here:
Convergence of a sequence with assumption that exponential subsequences converge?
1 answer
Sequence $a_n$ has a property that for every real $x > 1$, sequence $a_{lfloor{x^n}rfloor}$ converges to $0$. Does that mean that $a_n$ converges to $0$? I have tried to find a counterexample, but failed at it, so I think it might be true, but I don't know how to prove it.
This question already has an answer here:
Convergence of a sequence with assumption that exponential subsequences converge?
1 answer
real-analysis limits convergence
real-analysis limits convergence
edited Nov 27 at 9:50
asked Nov 27 at 8:34
Quo Si Than
1437
1437
marked as duplicate by gimusi
StackExchange.ready(function() {
if (StackExchange.options.isMobile) return;
$('.dupe-hammer-message-hover:not(.hover-bound)').each(function() {
var $hover = $(this).addClass('hover-bound'),
$msg = $hover.siblings('.dupe-hammer-message');
$hover.hover(
function() {
$hover.showInfoMessage('', {
messageElement: $msg.clone().show(),
transient: false,
position: { my: 'bottom left', at: 'top center', offsetTop: -7 },
dismissable: false,
relativeToBody: true
});
},
function() {
StackExchange.helpers.removeMessages();
}
);
});
});
Nov 28 at 8:08
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 gimusi
StackExchange.ready(function() {
if (StackExchange.options.isMobile) return;
$('.dupe-hammer-message-hover:not(.hover-bound)').each(function() {
var $hover = $(this).addClass('hover-bound'),
$msg = $hover.siblings('.dupe-hammer-message');
$hover.hover(
function() {
$hover.showInfoMessage('', {
messageElement: $msg.clone().show(),
transient: false,
position: { my: 'bottom left', at: 'top center', offsetTop: -7 },
dismissable: false,
relativeToBody: true
});
},
function() {
StackExchange.helpers.removeMessages();
}
);
});
});
Nov 28 at 8:08
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.
$a_n$ converges to 0 is equivalent to the following proposition : For every sequence $u_n$, there exists a real $x > 1$ such that ${u_n}_{n in mathbb{N}} cap {lfloor{x^n}rfloor}_{n in mathbb{N}}$ is infinite. I'm not sure how to procede from here.
– Rchn
Nov 27 at 13:26
For every *strictly increasing sequence $u_n$.
– Rchn
Nov 27 at 14:34
add a comment |
$a_n$ converges to 0 is equivalent to the following proposition : For every sequence $u_n$, there exists a real $x > 1$ such that ${u_n}_{n in mathbb{N}} cap {lfloor{x^n}rfloor}_{n in mathbb{N}}$ is infinite. I'm not sure how to procede from here.
– Rchn
Nov 27 at 13:26
For every *strictly increasing sequence $u_n$.
– Rchn
Nov 27 at 14:34
$a_n$ converges to 0 is equivalent to the following proposition : For every sequence $u_n$, there exists a real $x > 1$ such that ${u_n}_{n in mathbb{N}} cap {lfloor{x^n}rfloor}_{n in mathbb{N}}$ is infinite. I'm not sure how to procede from here.
– Rchn
Nov 27 at 13:26
$a_n$ converges to 0 is equivalent to the following proposition : For every sequence $u_n$, there exists a real $x > 1$ such that ${u_n}_{n in mathbb{N}} cap {lfloor{x^n}rfloor}_{n in mathbb{N}}$ is infinite. I'm not sure how to procede from here.
– Rchn
Nov 27 at 13:26
For every *strictly increasing sequence $u_n$.
– Rchn
Nov 27 at 14:34
For every *strictly increasing sequence $u_n$.
– Rchn
Nov 27 at 14:34
add a comment |
2 Answers
2
active
oldest
votes
We will use the following general observation.
Proposition. (Croft's Lemma) Let $U_1, U_2, cdots$ be subsets of $mathbb{R}$ such that the interior $mathring{U}_j$ satisfies $sup mathring{U}_j = infty$ for all $j geq 1$. Then the set
$$ mathcal{D} = { r in mathbb{R} : text{for each $j geq 1$, $n r in U_j$ holds for infinitely many $n$} } $$
is dense in $(0, infty)$.
Before proving this, let us rejoice its consequence to the problem of interest.
Claim. If $a_{lfloor x^n rfloor} to 0$ as $ntoinfty$ for each $x > 1$, then $a_n to 0$.
Proof. Define $f(r) = a_{lfloor exp(r) rfloor}$. Then the assumption tells that $f(nr) to 0$ as $ntoinfty$ along $mathbb{N}$, for each $r > 0$. Assume otherwise that $a_n notto 0$. Then there exists $epsilon > 0$ such that $|a_n| > epsilon$ holds for infinitely many $n$. Then
$$ U := { r : |f(r)| > epsilon } = bigcup_{n : |a_n| > epsilon} [log n, log(n+1)) $$
is such that its interior $mathring{U}$ is not bounded from above. So by Prpoposition, the set
$$ mathcal{D} = { r in (0, infty) : text{$nr in U$ for infinitely many $n$} } $$
is dense in $(0, infty)$, and in particular, non-empty. But if $r in mathcal{D}$, then $|f(nr)| > epsilon$ for infinitely many $n$, and so, $f(nr) notto 0$, a contradiction. ////
Lemma. If $U$ is an open subset of $mathbb{R}$ such that $sup U = infty$, then $bigcup_{ngeq N} frac{1}{n} U$ is dense in $(0, infty)$.
Proof. Write $mathcal{W}_N = bigcup_{ngeq N} frac{1}{n} U$. We aim to show that $(a, b) cap mathcal{W}_N$ is non-empty for each $0 < a < b$. Since $(n+1) a < n b$ for sufficiently large $n$, we can find $r > 0$ such that
$$(r, infty) subseteq bigcup_{ngeq N}(n a, n b).$$
Since $U$ is not bounded above, $U cap (r, infty)$ is non-empty, so we can pick $x in U cap (r, infty)$. Then $x in (n a, n b)$ for some $ngeq N$, and hence $frac{x}{n} in (a, b) cap frac{1}{n} U$. ////
Proof of Proposition. Notice that
$$ mathcal{D} = bigcap_{jgeq 1} bigcap_{Ngeq 1}bigcup_{ngeq N} frac{1}{n} U_j. $$
By the lemma, $mathcal{D}$ is a countable intersection of open dense subsets of $(0, infty)$. Then the conclusion follows from the Baire category theorem and the lemma above. ////
add a comment |
This is only a partial answer because I will assume that
$$
tag{C}forall xgt 1, quad lim_{nto +infty}a_{lfloor x^nrfloor +1}=0.
$$
I do not not whether this is a consequence of the original assumption on $(a_n)_n$.
Define a continuous function on $[1,+infty)$ in the following way. For each integer $ngeqslant 1$, we define $f(n):=a_n$ and we interpolate $f$ linearly on $(n,n+1)$. In particular, for $xin [n,n+1]]$, $f(x)$ belongs to the interval
$[min{a_n,a_{n+1}},max{a_n,a_{n+1}}}]$. It thus suffices to prove that $lim_{xto +infty}f(x)=0$.
Using (C) and the construction of $f$, we know that for all $xgt 1$, the sequence $left(fleft(x^nright)right)_{ngeqslant 1}$ goes to zero as $n$ goes to infinity. Define $g(x):=fleft(e^xright)$ for $xgeqslant 0$. Since $g(nx)to 0$ for all positive $x$, it follows that $g(x)to 0$ as $x$ goes to infinity hence so does $f$.
add a comment |
2 Answers
2
active
oldest
votes
2 Answers
2
active
oldest
votes
active
oldest
votes
active
oldest
votes
We will use the following general observation.
Proposition. (Croft's Lemma) Let $U_1, U_2, cdots$ be subsets of $mathbb{R}$ such that the interior $mathring{U}_j$ satisfies $sup mathring{U}_j = infty$ for all $j geq 1$. Then the set
$$ mathcal{D} = { r in mathbb{R} : text{for each $j geq 1$, $n r in U_j$ holds for infinitely many $n$} } $$
is dense in $(0, infty)$.
Before proving this, let us rejoice its consequence to the problem of interest.
Claim. If $a_{lfloor x^n rfloor} to 0$ as $ntoinfty$ for each $x > 1$, then $a_n to 0$.
Proof. Define $f(r) = a_{lfloor exp(r) rfloor}$. Then the assumption tells that $f(nr) to 0$ as $ntoinfty$ along $mathbb{N}$, for each $r > 0$. Assume otherwise that $a_n notto 0$. Then there exists $epsilon > 0$ such that $|a_n| > epsilon$ holds for infinitely many $n$. Then
$$ U := { r : |f(r)| > epsilon } = bigcup_{n : |a_n| > epsilon} [log n, log(n+1)) $$
is such that its interior $mathring{U}$ is not bounded from above. So by Prpoposition, the set
$$ mathcal{D} = { r in (0, infty) : text{$nr in U$ for infinitely many $n$} } $$
is dense in $(0, infty)$, and in particular, non-empty. But if $r in mathcal{D}$, then $|f(nr)| > epsilon$ for infinitely many $n$, and so, $f(nr) notto 0$, a contradiction. ////
Lemma. If $U$ is an open subset of $mathbb{R}$ such that $sup U = infty$, then $bigcup_{ngeq N} frac{1}{n} U$ is dense in $(0, infty)$.
Proof. Write $mathcal{W}_N = bigcup_{ngeq N} frac{1}{n} U$. We aim to show that $(a, b) cap mathcal{W}_N$ is non-empty for each $0 < a < b$. Since $(n+1) a < n b$ for sufficiently large $n$, we can find $r > 0$ such that
$$(r, infty) subseteq bigcup_{ngeq N}(n a, n b).$$
Since $U$ is not bounded above, $U cap (r, infty)$ is non-empty, so we can pick $x in U cap (r, infty)$. Then $x in (n a, n b)$ for some $ngeq N$, and hence $frac{x}{n} in (a, b) cap frac{1}{n} U$. ////
Proof of Proposition. Notice that
$$ mathcal{D} = bigcap_{jgeq 1} bigcap_{Ngeq 1}bigcup_{ngeq N} frac{1}{n} U_j. $$
By the lemma, $mathcal{D}$ is a countable intersection of open dense subsets of $(0, infty)$. Then the conclusion follows from the Baire category theorem and the lemma above. ////
add a comment |
We will use the following general observation.
Proposition. (Croft's Lemma) Let $U_1, U_2, cdots$ be subsets of $mathbb{R}$ such that the interior $mathring{U}_j$ satisfies $sup mathring{U}_j = infty$ for all $j geq 1$. Then the set
$$ mathcal{D} = { r in mathbb{R} : text{for each $j geq 1$, $n r in U_j$ holds for infinitely many $n$} } $$
is dense in $(0, infty)$.
Before proving this, let us rejoice its consequence to the problem of interest.
Claim. If $a_{lfloor x^n rfloor} to 0$ as $ntoinfty$ for each $x > 1$, then $a_n to 0$.
Proof. Define $f(r) = a_{lfloor exp(r) rfloor}$. Then the assumption tells that $f(nr) to 0$ as $ntoinfty$ along $mathbb{N}$, for each $r > 0$. Assume otherwise that $a_n notto 0$. Then there exists $epsilon > 0$ such that $|a_n| > epsilon$ holds for infinitely many $n$. Then
$$ U := { r : |f(r)| > epsilon } = bigcup_{n : |a_n| > epsilon} [log n, log(n+1)) $$
is such that its interior $mathring{U}$ is not bounded from above. So by Prpoposition, the set
$$ mathcal{D} = { r in (0, infty) : text{$nr in U$ for infinitely many $n$} } $$
is dense in $(0, infty)$, and in particular, non-empty. But if $r in mathcal{D}$, then $|f(nr)| > epsilon$ for infinitely many $n$, and so, $f(nr) notto 0$, a contradiction. ////
Lemma. If $U$ is an open subset of $mathbb{R}$ such that $sup U = infty$, then $bigcup_{ngeq N} frac{1}{n} U$ is dense in $(0, infty)$.
Proof. Write $mathcal{W}_N = bigcup_{ngeq N} frac{1}{n} U$. We aim to show that $(a, b) cap mathcal{W}_N$ is non-empty for each $0 < a < b$. Since $(n+1) a < n b$ for sufficiently large $n$, we can find $r > 0$ such that
$$(r, infty) subseteq bigcup_{ngeq N}(n a, n b).$$
Since $U$ is not bounded above, $U cap (r, infty)$ is non-empty, so we can pick $x in U cap (r, infty)$. Then $x in (n a, n b)$ for some $ngeq N$, and hence $frac{x}{n} in (a, b) cap frac{1}{n} U$. ////
Proof of Proposition. Notice that
$$ mathcal{D} = bigcap_{jgeq 1} bigcap_{Ngeq 1}bigcup_{ngeq N} frac{1}{n} U_j. $$
By the lemma, $mathcal{D}$ is a countable intersection of open dense subsets of $(0, infty)$. Then the conclusion follows from the Baire category theorem and the lemma above. ////
add a comment |
We will use the following general observation.
Proposition. (Croft's Lemma) Let $U_1, U_2, cdots$ be subsets of $mathbb{R}$ such that the interior $mathring{U}_j$ satisfies $sup mathring{U}_j = infty$ for all $j geq 1$. Then the set
$$ mathcal{D} = { r in mathbb{R} : text{for each $j geq 1$, $n r in U_j$ holds for infinitely many $n$} } $$
is dense in $(0, infty)$.
Before proving this, let us rejoice its consequence to the problem of interest.
Claim. If $a_{lfloor x^n rfloor} to 0$ as $ntoinfty$ for each $x > 1$, then $a_n to 0$.
Proof. Define $f(r) = a_{lfloor exp(r) rfloor}$. Then the assumption tells that $f(nr) to 0$ as $ntoinfty$ along $mathbb{N}$, for each $r > 0$. Assume otherwise that $a_n notto 0$. Then there exists $epsilon > 0$ such that $|a_n| > epsilon$ holds for infinitely many $n$. Then
$$ U := { r : |f(r)| > epsilon } = bigcup_{n : |a_n| > epsilon} [log n, log(n+1)) $$
is such that its interior $mathring{U}$ is not bounded from above. So by Prpoposition, the set
$$ mathcal{D} = { r in (0, infty) : text{$nr in U$ for infinitely many $n$} } $$
is dense in $(0, infty)$, and in particular, non-empty. But if $r in mathcal{D}$, then $|f(nr)| > epsilon$ for infinitely many $n$, and so, $f(nr) notto 0$, a contradiction. ////
Lemma. If $U$ is an open subset of $mathbb{R}$ such that $sup U = infty$, then $bigcup_{ngeq N} frac{1}{n} U$ is dense in $(0, infty)$.
Proof. Write $mathcal{W}_N = bigcup_{ngeq N} frac{1}{n} U$. We aim to show that $(a, b) cap mathcal{W}_N$ is non-empty for each $0 < a < b$. Since $(n+1) a < n b$ for sufficiently large $n$, we can find $r > 0$ such that
$$(r, infty) subseteq bigcup_{ngeq N}(n a, n b).$$
Since $U$ is not bounded above, $U cap (r, infty)$ is non-empty, so we can pick $x in U cap (r, infty)$. Then $x in (n a, n b)$ for some $ngeq N$, and hence $frac{x}{n} in (a, b) cap frac{1}{n} U$. ////
Proof of Proposition. Notice that
$$ mathcal{D} = bigcap_{jgeq 1} bigcap_{Ngeq 1}bigcup_{ngeq N} frac{1}{n} U_j. $$
By the lemma, $mathcal{D}$ is a countable intersection of open dense subsets of $(0, infty)$. Then the conclusion follows from the Baire category theorem and the lemma above. ////
We will use the following general observation.
Proposition. (Croft's Lemma) Let $U_1, U_2, cdots$ be subsets of $mathbb{R}$ such that the interior $mathring{U}_j$ satisfies $sup mathring{U}_j = infty$ for all $j geq 1$. Then the set
$$ mathcal{D} = { r in mathbb{R} : text{for each $j geq 1$, $n r in U_j$ holds for infinitely many $n$} } $$
is dense in $(0, infty)$.
Before proving this, let us rejoice its consequence to the problem of interest.
Claim. If $a_{lfloor x^n rfloor} to 0$ as $ntoinfty$ for each $x > 1$, then $a_n to 0$.
Proof. Define $f(r) = a_{lfloor exp(r) rfloor}$. Then the assumption tells that $f(nr) to 0$ as $ntoinfty$ along $mathbb{N}$, for each $r > 0$. Assume otherwise that $a_n notto 0$. Then there exists $epsilon > 0$ such that $|a_n| > epsilon$ holds for infinitely many $n$. Then
$$ U := { r : |f(r)| > epsilon } = bigcup_{n : |a_n| > epsilon} [log n, log(n+1)) $$
is such that its interior $mathring{U}$ is not bounded from above. So by Prpoposition, the set
$$ mathcal{D} = { r in (0, infty) : text{$nr in U$ for infinitely many $n$} } $$
is dense in $(0, infty)$, and in particular, non-empty. But if $r in mathcal{D}$, then $|f(nr)| > epsilon$ for infinitely many $n$, and so, $f(nr) notto 0$, a contradiction. ////
Lemma. If $U$ is an open subset of $mathbb{R}$ such that $sup U = infty$, then $bigcup_{ngeq N} frac{1}{n} U$ is dense in $(0, infty)$.
Proof. Write $mathcal{W}_N = bigcup_{ngeq N} frac{1}{n} U$. We aim to show that $(a, b) cap mathcal{W}_N$ is non-empty for each $0 < a < b$. Since $(n+1) a < n b$ for sufficiently large $n$, we can find $r > 0$ such that
$$(r, infty) subseteq bigcup_{ngeq N}(n a, n b).$$
Since $U$ is not bounded above, $U cap (r, infty)$ is non-empty, so we can pick $x in U cap (r, infty)$. Then $x in (n a, n b)$ for some $ngeq N$, and hence $frac{x}{n} in (a, b) cap frac{1}{n} U$. ////
Proof of Proposition. Notice that
$$ mathcal{D} = bigcap_{jgeq 1} bigcap_{Ngeq 1}bigcup_{ngeq N} frac{1}{n} U_j. $$
By the lemma, $mathcal{D}$ is a countable intersection of open dense subsets of $(0, infty)$. Then the conclusion follows from the Baire category theorem and the lemma above. ////
answered Nov 28 at 0:48
Sangchul Lee
91.2k12163264
91.2k12163264
add a comment |
add a comment |
This is only a partial answer because I will assume that
$$
tag{C}forall xgt 1, quad lim_{nto +infty}a_{lfloor x^nrfloor +1}=0.
$$
I do not not whether this is a consequence of the original assumption on $(a_n)_n$.
Define a continuous function on $[1,+infty)$ in the following way. For each integer $ngeqslant 1$, we define $f(n):=a_n$ and we interpolate $f$ linearly on $(n,n+1)$. In particular, for $xin [n,n+1]]$, $f(x)$ belongs to the interval
$[min{a_n,a_{n+1}},max{a_n,a_{n+1}}}]$. It thus suffices to prove that $lim_{xto +infty}f(x)=0$.
Using (C) and the construction of $f$, we know that for all $xgt 1$, the sequence $left(fleft(x^nright)right)_{ngeqslant 1}$ goes to zero as $n$ goes to infinity. Define $g(x):=fleft(e^xright)$ for $xgeqslant 0$. Since $g(nx)to 0$ for all positive $x$, it follows that $g(x)to 0$ as $x$ goes to infinity hence so does $f$.
add a comment |
This is only a partial answer because I will assume that
$$
tag{C}forall xgt 1, quad lim_{nto +infty}a_{lfloor x^nrfloor +1}=0.
$$
I do not not whether this is a consequence of the original assumption on $(a_n)_n$.
Define a continuous function on $[1,+infty)$ in the following way. For each integer $ngeqslant 1$, we define $f(n):=a_n$ and we interpolate $f$ linearly on $(n,n+1)$. In particular, for $xin [n,n+1]]$, $f(x)$ belongs to the interval
$[min{a_n,a_{n+1}},max{a_n,a_{n+1}}}]$. It thus suffices to prove that $lim_{xto +infty}f(x)=0$.
Using (C) and the construction of $f$, we know that for all $xgt 1$, the sequence $left(fleft(x^nright)right)_{ngeqslant 1}$ goes to zero as $n$ goes to infinity. Define $g(x):=fleft(e^xright)$ for $xgeqslant 0$. Since $g(nx)to 0$ for all positive $x$, it follows that $g(x)to 0$ as $x$ goes to infinity hence so does $f$.
add a comment |
This is only a partial answer because I will assume that
$$
tag{C}forall xgt 1, quad lim_{nto +infty}a_{lfloor x^nrfloor +1}=0.
$$
I do not not whether this is a consequence of the original assumption on $(a_n)_n$.
Define a continuous function on $[1,+infty)$ in the following way. For each integer $ngeqslant 1$, we define $f(n):=a_n$ and we interpolate $f$ linearly on $(n,n+1)$. In particular, for $xin [n,n+1]]$, $f(x)$ belongs to the interval
$[min{a_n,a_{n+1}},max{a_n,a_{n+1}}}]$. It thus suffices to prove that $lim_{xto +infty}f(x)=0$.
Using (C) and the construction of $f$, we know that for all $xgt 1$, the sequence $left(fleft(x^nright)right)_{ngeqslant 1}$ goes to zero as $n$ goes to infinity. Define $g(x):=fleft(e^xright)$ for $xgeqslant 0$. Since $g(nx)to 0$ for all positive $x$, it follows that $g(x)to 0$ as $x$ goes to infinity hence so does $f$.
This is only a partial answer because I will assume that
$$
tag{C}forall xgt 1, quad lim_{nto +infty}a_{lfloor x^nrfloor +1}=0.
$$
I do not not whether this is a consequence of the original assumption on $(a_n)_n$.
Define a continuous function on $[1,+infty)$ in the following way. For each integer $ngeqslant 1$, we define $f(n):=a_n$ and we interpolate $f$ linearly on $(n,n+1)$. In particular, for $xin [n,n+1]]$, $f(x)$ belongs to the interval
$[min{a_n,a_{n+1}},max{a_n,a_{n+1}}}]$. It thus suffices to prove that $lim_{xto +infty}f(x)=0$.
Using (C) and the construction of $f$, we know that for all $xgt 1$, the sequence $left(fleft(x^nright)right)_{ngeqslant 1}$ goes to zero as $n$ goes to infinity. Define $g(x):=fleft(e^xright)$ for $xgeqslant 0$. Since $g(nx)to 0$ for all positive $x$, it follows that $g(x)to 0$ as $x$ goes to infinity hence so does $f$.
answered Nov 27 at 15:23
Davide Giraudo
125k16150259
125k16150259
add a comment |
add a comment |
$a_n$ converges to 0 is equivalent to the following proposition : For every sequence $u_n$, there exists a real $x > 1$ such that ${u_n}_{n in mathbb{N}} cap {lfloor{x^n}rfloor}_{n in mathbb{N}}$ is infinite. I'm not sure how to procede from here.
– Rchn
Nov 27 at 13:26
For every *strictly increasing sequence $u_n$.
– Rchn
Nov 27 at 14:34