Can AM-GM inequality prove 2^n > 1+n\sqrt{2^{n-1}} for positive integers?

  • Thread starter Thread starter utkarshakash
  • Start date Start date
  • Tags Tags
    Inequality
AI Thread Summary
The discussion revolves around proving the inequality 2^n > 1 + n√(2^(n-1)) for positive integers n. Participants suggest using the AM-GM inequality but express uncertainty about the appropriate numbers to apply it to. There is a debate on whether the inequality holds for all positive integers, with examples showing it may not be true for n=1. Mathematical induction is recommended as a more effective method for proving the statement, with hints provided on how to approach the proof using the binomial theorem and properties of arithmetic and geometric means. Ultimately, the consensus leans towards induction as the preferred technique for this proof.
utkarshakash
Gold Member
Messages
852
Reaction score
13

Homework Statement


If n is a positive integer, prove that 2^n > 1+n\sqrt{2^{n-1}}

Homework Equations



The Attempt at a Solution


I am thinking of applying AM GM HM inequality. But which numbers should I take to arrive at this inequality?
 
Physics news on Phys.org
utkarshakash said:

Homework Statement


If n is a positive integer, prove that 2^n > 1+n\sqrt{2^{n-1}}

Homework Equations



The Attempt at a Solution


I am thinking of applying AM GM HM inequality. But which numbers should I take to arrive at this inequality?

...This may not be correct for any number n that is a positive integer.

For example, 2^{1} is not greater than 1 + n\sqrt{2^{1-1}}. In fact, they are equivalent.
 
Mandelbroth said:
...This may not be correct for any number n that is a positive integer.

For example, 2^{1} is not greater than 1 + n\sqrt{2^{1-1}}. In fact, they are equivalent.

Ok so assume that it is 'greater that or equal to' instead of just 'is greater than' and prove it
 
utkarshakash said:
Ok so assume that it is 'greater that or equal to' instead of just 'is greater than' and prove it

No: YOU prove it, or at least show some effort towards the solution. Read the Forum rules!

RGV
 
Ray Vickson said:
No: YOU prove it, or at least show some effort towards the solution. Read the Forum rules!

RGV

Hey I really don't know how to solve this. I need some hints to get started. I've already stated that I am thinking of solving it using AM GM HM inequality. I know nothing more than this.
 
utkarshakash said:
Hey I really don't know how to solve this. I need some hints to get started. I've already stated that I am thinking of solving it using AM GM HM inequality. I know nothing more than this.

Use mathematical Induction. Do you know what is it ?
 
I think appying the binomial theorem to (1+ 1)n would also work.
 
sankalpmittal said:
Use mathematical Induction. Do you know what is it ?

Yes I know but I'm not required to use it. Also I'm not good at it.
 
utkarshakash said:
Yes I know but I'm not required to use it. Also I'm not good at it.

This might be your chance to improve on your technique then. Problems like this, beginning with "Prove for any n...", practically yell out "Use induction on me!" And mostly it's the easiest way to solve them.

I can see the binomial theorem being used to start 2n = (1+1)n = 1 + n Ʃ..., but getting from that sum to √2n-1 may be more work that induction would be. Maybe I'm overlooking some obvious trick though.
 
  • #10
utkarshakash said:
Yes I know but I'm not required to use it. Also I'm not good at it.

I understand what you want... But believe instead of using "Arithematic mean ≥ Geometric mean" , its more easier to use mathematical induction anyways...

But you want to use progression and series only... Ok , well , I throw you off a hint :

2n =2x2n-1.Now use A.M≥G.M here..
 
Back
Top