Prove that if $sum u_n$ is convergent, then $sumsqrt[n]n,u_n$ is also convergent [closed]
Clash Royale CLAN TAG#URR8PPP
up vote
1
down vote
favorite
If $sum u_n$ is a convergent series then can it be concluded that
$sumsqrt[n]n,u_n$ is also convergent
I think that by abels theorem since $sqrt[n]n$ is monotone bounded hence the above theorem holds true
Please help from there
sequences-and-series
closed as off-topic by flawr, Siong Thye Goh, Nosrati, amWhy, stressed out Aug 12 at 20:41
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." â Nosrati, amWhy, stressed out
add a comment |Â
up vote
1
down vote
favorite
If $sum u_n$ is a convergent series then can it be concluded that
$sumsqrt[n]n,u_n$ is also convergent
I think that by abels theorem since $sqrt[n]n$ is monotone bounded hence the above theorem holds true
Please help from there
sequences-and-series
closed as off-topic by flawr, Siong Thye Goh, Nosrati, amWhy, stressed out Aug 12 at 20:41
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." â Nosrati, amWhy, stressed out
Yes it is a consequence of Abel's theorem, but not applied to $n^1/n$ since it is not monotone. You need to skip a finite number of terms from the beginning since what it really is is eventually decreasing and bounded by $1$ from below.
â user583012
Aug 11 at 14:47
Please also show the index: $nge?$
â Szeto
Aug 11 at 15:21
2
It is more often called an "Abel Test," not Abel's Theorem, which is different.
â Mark Viola
Aug 11 at 16:45
add a comment |Â
up vote
1
down vote
favorite
up vote
1
down vote
favorite
If $sum u_n$ is a convergent series then can it be concluded that
$sumsqrt[n]n,u_n$ is also convergent
I think that by abels theorem since $sqrt[n]n$ is monotone bounded hence the above theorem holds true
Please help from there
sequences-and-series
If $sum u_n$ is a convergent series then can it be concluded that
$sumsqrt[n]n,u_n$ is also convergent
I think that by abels theorem since $sqrt[n]n$ is monotone bounded hence the above theorem holds true
Please help from there
sequences-and-series
edited Aug 12 at 7:11
Sil
5,14121443
5,14121443
asked Aug 11 at 14:36
Soham Mukherjee
92
92
closed as off-topic by flawr, Siong Thye Goh, Nosrati, amWhy, stressed out Aug 12 at 20:41
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." â Nosrati, amWhy, stressed out
closed as off-topic by flawr, Siong Thye Goh, Nosrati, amWhy, stressed out Aug 12 at 20:41
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." â Nosrati, amWhy, stressed out
Yes it is a consequence of Abel's theorem, but not applied to $n^1/n$ since it is not monotone. You need to skip a finite number of terms from the beginning since what it really is is eventually decreasing and bounded by $1$ from below.
â user583012
Aug 11 at 14:47
Please also show the index: $nge?$
â Szeto
Aug 11 at 15:21
2
It is more often called an "Abel Test," not Abel's Theorem, which is different.
â Mark Viola
Aug 11 at 16:45
add a comment |Â
Yes it is a consequence of Abel's theorem, but not applied to $n^1/n$ since it is not monotone. You need to skip a finite number of terms from the beginning since what it really is is eventually decreasing and bounded by $1$ from below.
â user583012
Aug 11 at 14:47
Please also show the index: $nge?$
â Szeto
Aug 11 at 15:21
2
It is more often called an "Abel Test," not Abel's Theorem, which is different.
â Mark Viola
Aug 11 at 16:45
Yes it is a consequence of Abel's theorem, but not applied to $n^1/n$ since it is not monotone. You need to skip a finite number of terms from the beginning since what it really is is eventually decreasing and bounded by $1$ from below.
â user583012
Aug 11 at 14:47
Yes it is a consequence of Abel's theorem, but not applied to $n^1/n$ since it is not monotone. You need to skip a finite number of terms from the beginning since what it really is is eventually decreasing and bounded by $1$ from below.
â user583012
Aug 11 at 14:47
Please also show the index: $nge?$
â Szeto
Aug 11 at 15:21
Please also show the index: $nge?$
â Szeto
Aug 11 at 15:21
2
2
It is more often called an "Abel Test," not Abel's Theorem, which is different.
â Mark Viola
Aug 11 at 16:45
It is more often called an "Abel Test," not Abel's Theorem, which is different.
â Mark Viola
Aug 11 at 16:45
add a comment |Â
2 Answers
2
active
oldest
votes
up vote
2
down vote
If $u_n geq 0$ for all sufficiently large $n$, yes. And this is easy.
Proof: without loss of generality, assume $u_n > 0$ for all $n$.
$
lim_nto infty sqrt[n]n = 1
$, so $sqrt[n]nu_n sim_ntoinftyu_n$ and you can conclude by the limit comparison test.If $(u_n)_n$ is allowed arbitrary signs, yes. But this is less obvious (though not horrendous either), and as you suggested follows from Abel/Dirichlet's test.
(What follows is not the shortest proof, I reckon, but I find it instructive.)
Note that since $$sqrt[n]n = e^fracln nn = 1+ fracln nn+Oleft(fracln^2 nn^2right)tag1$$
when $nto infty$, and $lim_ntoinftyu_n = 0$ we have
$$
sqrt[n]nu_n = u_n + fracln nnu_n + varepsilon_n tag2.
$$
with $varepsilon_n = oleft(fracln^2 nn^2right)$.
By assumption, $sum_n u_n$ is a convergent series. By comparison with a $p$-series, the series $sum_n varepsilon_n$ is absolutely convergent. So overall, we have that
$$
sum_n sqrt[n]nu_n
$$
converges if, and only if,
$$
sum_n fracln nnu_n
$$
does. So it suffices to prove the latter: this is where Dirichlet's test comes in handy: we apply it with $a_n = fraclnn$, $b_n=u_n$, so that (i) $(a_n)_n$ is non-increasing with limit $0$; (ii) the partial sums of $sum_n b_n$ are uniformly bounded (since $sum_n u_n$ is convergent. $square$
add a comment |Â
up vote
-1
down vote
$$fraca_n+1a_n=frac(n+1)^frac1n+1 u_n+1(n)^frac1n u_n$$
$$fraca_n+1a_nlefrac(n+1)^frac1n u_n+1(n)^frac1n u_n $$
$$fraca_n+1a_nlefrac(fracn+1n)^frac1n u_n+1 u_n $$
$$fraca_n+1a_nlefrac(1+frac1n)^frac1n u_n+1 u_n $$
$u_n$ is convergent also $(1+1/n)^1/n$
1
The signs of $u_n$ are not known.
â user583012
Aug 11 at 14:56
2
This is not how we do ratio test formally. You missed the absolute value and limit. Also, what if $lim_ntoinfty|u_n+1/u_n|=1$? Then your test shows nothing at all!
â Szeto
Aug 11 at 15:12
In addition, if you do ratio test correctly, you still get $lim_ntoinfty|a_n+1/a_n|le 1$ which is not sufficient to prove convergence.
â Szeto
Aug 11 at 15:14
1
@upvoter Care to explain your vote?
â Did
Aug 12 at 9:11
add a comment |Â
2 Answers
2
active
oldest
votes
2 Answers
2
active
oldest
votes
active
oldest
votes
active
oldest
votes
up vote
2
down vote
If $u_n geq 0$ for all sufficiently large $n$, yes. And this is easy.
Proof: without loss of generality, assume $u_n > 0$ for all $n$.
$
lim_nto infty sqrt[n]n = 1
$, so $sqrt[n]nu_n sim_ntoinftyu_n$ and you can conclude by the limit comparison test.If $(u_n)_n$ is allowed arbitrary signs, yes. But this is less obvious (though not horrendous either), and as you suggested follows from Abel/Dirichlet's test.
(What follows is not the shortest proof, I reckon, but I find it instructive.)
Note that since $$sqrt[n]n = e^fracln nn = 1+ fracln nn+Oleft(fracln^2 nn^2right)tag1$$
when $nto infty$, and $lim_ntoinftyu_n = 0$ we have
$$
sqrt[n]nu_n = u_n + fracln nnu_n + varepsilon_n tag2.
$$
with $varepsilon_n = oleft(fracln^2 nn^2right)$.
By assumption, $sum_n u_n$ is a convergent series. By comparison with a $p$-series, the series $sum_n varepsilon_n$ is absolutely convergent. So overall, we have that
$$
sum_n sqrt[n]nu_n
$$
converges if, and only if,
$$
sum_n fracln nnu_n
$$
does. So it suffices to prove the latter: this is where Dirichlet's test comes in handy: we apply it with $a_n = fraclnn$, $b_n=u_n$, so that (i) $(a_n)_n$ is non-increasing with limit $0$; (ii) the partial sums of $sum_n b_n$ are uniformly bounded (since $sum_n u_n$ is convergent. $square$
add a comment |Â
up vote
2
down vote
If $u_n geq 0$ for all sufficiently large $n$, yes. And this is easy.
Proof: without loss of generality, assume $u_n > 0$ for all $n$.
$
lim_nto infty sqrt[n]n = 1
$, so $sqrt[n]nu_n sim_ntoinftyu_n$ and you can conclude by the limit comparison test.If $(u_n)_n$ is allowed arbitrary signs, yes. But this is less obvious (though not horrendous either), and as you suggested follows from Abel/Dirichlet's test.
(What follows is not the shortest proof, I reckon, but I find it instructive.)
Note that since $$sqrt[n]n = e^fracln nn = 1+ fracln nn+Oleft(fracln^2 nn^2right)tag1$$
when $nto infty$, and $lim_ntoinftyu_n = 0$ we have
$$
sqrt[n]nu_n = u_n + fracln nnu_n + varepsilon_n tag2.
$$
with $varepsilon_n = oleft(fracln^2 nn^2right)$.
By assumption, $sum_n u_n$ is a convergent series. By comparison with a $p$-series, the series $sum_n varepsilon_n$ is absolutely convergent. So overall, we have that
$$
sum_n sqrt[n]nu_n
$$
converges if, and only if,
$$
sum_n fracln nnu_n
$$
does. So it suffices to prove the latter: this is where Dirichlet's test comes in handy: we apply it with $a_n = fraclnn$, $b_n=u_n$, so that (i) $(a_n)_n$ is non-increasing with limit $0$; (ii) the partial sums of $sum_n b_n$ are uniformly bounded (since $sum_n u_n$ is convergent. $square$
add a comment |Â
up vote
2
down vote
up vote
2
down vote
If $u_n geq 0$ for all sufficiently large $n$, yes. And this is easy.
Proof: without loss of generality, assume $u_n > 0$ for all $n$.
$
lim_nto infty sqrt[n]n = 1
$, so $sqrt[n]nu_n sim_ntoinftyu_n$ and you can conclude by the limit comparison test.If $(u_n)_n$ is allowed arbitrary signs, yes. But this is less obvious (though not horrendous either), and as you suggested follows from Abel/Dirichlet's test.
(What follows is not the shortest proof, I reckon, but I find it instructive.)
Note that since $$sqrt[n]n = e^fracln nn = 1+ fracln nn+Oleft(fracln^2 nn^2right)tag1$$
when $nto infty$, and $lim_ntoinftyu_n = 0$ we have
$$
sqrt[n]nu_n = u_n + fracln nnu_n + varepsilon_n tag2.
$$
with $varepsilon_n = oleft(fracln^2 nn^2right)$.
By assumption, $sum_n u_n$ is a convergent series. By comparison with a $p$-series, the series $sum_n varepsilon_n$ is absolutely convergent. So overall, we have that
$$
sum_n sqrt[n]nu_n
$$
converges if, and only if,
$$
sum_n fracln nnu_n
$$
does. So it suffices to prove the latter: this is where Dirichlet's test comes in handy: we apply it with $a_n = fraclnn$, $b_n=u_n$, so that (i) $(a_n)_n$ is non-increasing with limit $0$; (ii) the partial sums of $sum_n b_n$ are uniformly bounded (since $sum_n u_n$ is convergent. $square$
If $u_n geq 0$ for all sufficiently large $n$, yes. And this is easy.
Proof: without loss of generality, assume $u_n > 0$ for all $n$.
$
lim_nto infty sqrt[n]n = 1
$, so $sqrt[n]nu_n sim_ntoinftyu_n$ and you can conclude by the limit comparison test.If $(u_n)_n$ is allowed arbitrary signs, yes. But this is less obvious (though not horrendous either), and as you suggested follows from Abel/Dirichlet's test.
(What follows is not the shortest proof, I reckon, but I find it instructive.)
Note that since $$sqrt[n]n = e^fracln nn = 1+ fracln nn+Oleft(fracln^2 nn^2right)tag1$$
when $nto infty$, and $lim_ntoinftyu_n = 0$ we have
$$
sqrt[n]nu_n = u_n + fracln nnu_n + varepsilon_n tag2.
$$
with $varepsilon_n = oleft(fracln^2 nn^2right)$.
By assumption, $sum_n u_n$ is a convergent series. By comparison with a $p$-series, the series $sum_n varepsilon_n$ is absolutely convergent. So overall, we have that
$$
sum_n sqrt[n]nu_n
$$
converges if, and only if,
$$
sum_n fracln nnu_n
$$
does. So it suffices to prove the latter: this is where Dirichlet's test comes in handy: we apply it with $a_n = fraclnn$, $b_n=u_n$, so that (i) $(a_n)_n$ is non-increasing with limit $0$; (ii) the partial sums of $sum_n b_n$ are uniformly bounded (since $sum_n u_n$ is convergent. $square$
answered Aug 11 at 16:55
Clement C.
47.2k33682
47.2k33682
add a comment |Â
add a comment |Â
up vote
-1
down vote
$$fraca_n+1a_n=frac(n+1)^frac1n+1 u_n+1(n)^frac1n u_n$$
$$fraca_n+1a_nlefrac(n+1)^frac1n u_n+1(n)^frac1n u_n $$
$$fraca_n+1a_nlefrac(fracn+1n)^frac1n u_n+1 u_n $$
$$fraca_n+1a_nlefrac(1+frac1n)^frac1n u_n+1 u_n $$
$u_n$ is convergent also $(1+1/n)^1/n$
1
The signs of $u_n$ are not known.
â user583012
Aug 11 at 14:56
2
This is not how we do ratio test formally. You missed the absolute value and limit. Also, what if $lim_ntoinfty|u_n+1/u_n|=1$? Then your test shows nothing at all!
â Szeto
Aug 11 at 15:12
In addition, if you do ratio test correctly, you still get $lim_ntoinfty|a_n+1/a_n|le 1$ which is not sufficient to prove convergence.
â Szeto
Aug 11 at 15:14
1
@upvoter Care to explain your vote?
â Did
Aug 12 at 9:11
add a comment |Â
up vote
-1
down vote
$$fraca_n+1a_n=frac(n+1)^frac1n+1 u_n+1(n)^frac1n u_n$$
$$fraca_n+1a_nlefrac(n+1)^frac1n u_n+1(n)^frac1n u_n $$
$$fraca_n+1a_nlefrac(fracn+1n)^frac1n u_n+1 u_n $$
$$fraca_n+1a_nlefrac(1+frac1n)^frac1n u_n+1 u_n $$
$u_n$ is convergent also $(1+1/n)^1/n$
1
The signs of $u_n$ are not known.
â user583012
Aug 11 at 14:56
2
This is not how we do ratio test formally. You missed the absolute value and limit. Also, what if $lim_ntoinfty|u_n+1/u_n|=1$? Then your test shows nothing at all!
â Szeto
Aug 11 at 15:12
In addition, if you do ratio test correctly, you still get $lim_ntoinfty|a_n+1/a_n|le 1$ which is not sufficient to prove convergence.
â Szeto
Aug 11 at 15:14
1
@upvoter Care to explain your vote?
â Did
Aug 12 at 9:11
add a comment |Â
up vote
-1
down vote
up vote
-1
down vote
$$fraca_n+1a_n=frac(n+1)^frac1n+1 u_n+1(n)^frac1n u_n$$
$$fraca_n+1a_nlefrac(n+1)^frac1n u_n+1(n)^frac1n u_n $$
$$fraca_n+1a_nlefrac(fracn+1n)^frac1n u_n+1 u_n $$
$$fraca_n+1a_nlefrac(1+frac1n)^frac1n u_n+1 u_n $$
$u_n$ is convergent also $(1+1/n)^1/n$
$$fraca_n+1a_n=frac(n+1)^frac1n+1 u_n+1(n)^frac1n u_n$$
$$fraca_n+1a_nlefrac(n+1)^frac1n u_n+1(n)^frac1n u_n $$
$$fraca_n+1a_nlefrac(fracn+1n)^frac1n u_n+1 u_n $$
$$fraca_n+1a_nlefrac(1+frac1n)^frac1n u_n+1 u_n $$
$u_n$ is convergent also $(1+1/n)^1/n$
answered Aug 11 at 14:51
James
1,803419
1,803419
1
The signs of $u_n$ are not known.
â user583012
Aug 11 at 14:56
2
This is not how we do ratio test formally. You missed the absolute value and limit. Also, what if $lim_ntoinfty|u_n+1/u_n|=1$? Then your test shows nothing at all!
â Szeto
Aug 11 at 15:12
In addition, if you do ratio test correctly, you still get $lim_ntoinfty|a_n+1/a_n|le 1$ which is not sufficient to prove convergence.
â Szeto
Aug 11 at 15:14
1
@upvoter Care to explain your vote?
â Did
Aug 12 at 9:11
add a comment |Â
1
The signs of $u_n$ are not known.
â user583012
Aug 11 at 14:56
2
This is not how we do ratio test formally. You missed the absolute value and limit. Also, what if $lim_ntoinfty|u_n+1/u_n|=1$? Then your test shows nothing at all!
â Szeto
Aug 11 at 15:12
In addition, if you do ratio test correctly, you still get $lim_ntoinfty|a_n+1/a_n|le 1$ which is not sufficient to prove convergence.
â Szeto
Aug 11 at 15:14
1
@upvoter Care to explain your vote?
â Did
Aug 12 at 9:11
1
1
The signs of $u_n$ are not known.
â user583012
Aug 11 at 14:56
The signs of $u_n$ are not known.
â user583012
Aug 11 at 14:56
2
2
This is not how we do ratio test formally. You missed the absolute value and limit. Also, what if $lim_ntoinfty|u_n+1/u_n|=1$? Then your test shows nothing at all!
â Szeto
Aug 11 at 15:12
This is not how we do ratio test formally. You missed the absolute value and limit. Also, what if $lim_ntoinfty|u_n+1/u_n|=1$? Then your test shows nothing at all!
â Szeto
Aug 11 at 15:12
In addition, if you do ratio test correctly, you still get $lim_ntoinfty|a_n+1/a_n|le 1$ which is not sufficient to prove convergence.
â Szeto
Aug 11 at 15:14
In addition, if you do ratio test correctly, you still get $lim_ntoinfty|a_n+1/a_n|le 1$ which is not sufficient to prove convergence.
â Szeto
Aug 11 at 15:14
1
1
@upvoter Care to explain your vote?
â Did
Aug 12 at 9:11
@upvoter Care to explain your vote?
â Did
Aug 12 at 9:11
add a comment |Â
Yes it is a consequence of Abel's theorem, but not applied to $n^1/n$ since it is not monotone. You need to skip a finite number of terms from the beginning since what it really is is eventually decreasing and bounded by $1$ from below.
â user583012
Aug 11 at 14:47
Please also show the index: $nge?$
â Szeto
Aug 11 at 15:21
2
It is more often called an "Abel Test," not Abel's Theorem, which is different.
â Mark Viola
Aug 11 at 16:45