Waste less time on Facebook — follow Brilliant.
×

Ideas Needed !

Is there a better way to find the following sum ? I mean is there a better approximation possible ?

\[\sum_{n=1}^{\infty} \sqrt {1+n^4} -n^2\]

A possible way could be :

\[ S = \displaystyle \sum_{n=1}^{\infty} \sqrt{1 + n^{4}} - n^{2} \\ S = \displaystyle \sum_{n=1}^{\infty} \frac { \left ( \sqrt{1 + n^{4}} - n^{2} \right ) \left ( \sqrt{1 + n^{4}} + n^{2} \right ) }{ \sqrt{1 + n^{4}} + n^{2} } \\= \displaystyle \sum_{n=1}^{\infty} \frac { 1 } { \sqrt{1 + n^{4}} + n^{2} } \\\\ \text{Now we can make an approximation that } \\\\S < \displaystyle \sum_{n=1}^{\infty} \frac { 1 } { \sqrt{ n^{4}} + n^{2} } \\ S < \displaystyle \sum_{n=1}^{\infty} \frac { 1 } { 2 n^{2} } \\ S < \displaystyle {\frac { \pi^{2}}{12} } \]

Note by Azhaghu Roopesh M
1 year, 10 months ago

No vote yet
1 vote

Comments

Sort by:

Top Newest

@Akshay Bodhare So you are back ? Azhaghu Roopesh M · 1 year, 10 months ago

Log in to reply

@Azhaghu Roopesh M Sorry for replying after a long time,

\(\frac{\pi^2}{12} \approx 0.8224\)

This bound is not tight enough,I suggest that you should use Cauchy-Schwarz inequality in last step to get the better approximation,

\(S\approx\dfrac{\pi \coth {\frac{\pi}{\sqrt{1+\sqrt{2}}}}-\sqrt{1+\sqrt{2}}}{\sqrt{2(1+\sqrt{2})}}\approx 0.7736211057\)

Btw,the sum coverges slowly

The approximate value of the sum according to CAS is,

\(S \approx 0.7345725812\) Akshay Bodhare · 1 year, 10 months ago

Log in to reply

@Akshay Bodhare Nice , I wonder why it didn't strike me ! +1

Also I'll try to think up a question based on it . Azhaghu Roopesh M · 1 year, 10 months ago

Log in to reply

@Azhaghu Roopesh M However,even the upper bound I gave is not tight enough,do you have any ideas for better approximation? Akshay Bodhare · 1 year, 10 months ago

Log in to reply

@Akshay Bodhare I think that I'll have to search on the I'net for it then , since my knowledge on it is limited .

I'll discuss with some of my other friends and I'll definitely get back to you :) Azhaghu Roopesh M · 1 year, 10 months ago

Log in to reply

@Azhaghu Roopesh M Could you please give me some hint(not solution) for this? I want to solve it!But if you think that I am violating any guidelines of Brilliant , you can delete this comment. Harsh Shrivastava · 1 year, 10 months ago

Log in to reply

@Harsh Shrivastava I'm sure you must have heard about Cauchy Schwartz inequality , haven't you ?

Also I can't delete any comment , I'm not a moderator you see ! Azhaghu Roopesh M · 1 year, 10 months ago

Log in to reply

@Azhaghu Roopesh M Hmmmm... I will try using it,

I thought you were a moderator, Sorry to bother you. Harsh Shrivastava · 1 year, 10 months ago

Log in to reply

@Harsh Shrivastava No problem :) Azhaghu Roopesh M · 1 year, 10 months ago

Log in to reply

Log in to reply

If you just want a better approximation, just calculate the sum of first few terms by hand, and use approximation for the rest. Raghav Vaidyanathan · 1 year, 10 months ago

Log in to reply

@Raghav Vaidyanathan I guess that'll do it but ok I'll try Azhaghu Roopesh M · 1 year, 10 months ago

Log in to reply

×

Problem Loading...

Note Loading...

Set Loading...