Prove False: n\geq a\Rightarrow n!\geq a^n

  • Level: Graduate 
  • Thread starter Thread starter 3.1415926535
  • Start date Start date
  • Tags Tags
    Induction
Join the discussion
Registration is free. Start your own thread to ask a follow-up.
5 replies · 4K views
3.1415926535
Messages
80
Reaction score
0
I will prove the false statement, that [tex]n\geq a\Rightarrow n!\geq a^n, n\in \mathbb{N}-\left \{ 0 \right \}[/tex] with induction

For [tex]n=1[/tex] [tex]1\geq a\Rightarrow 1!\geq a^1\Rightarrow 1 \geq a[/tex] which is true.

Suppose that [tex]n\geq a\Rightarrow n!\geq a^n, n\in \mathbb{N}-\left \{ 0 \right \}[/tex]

Then,
[tex]n\geq a\Rightarrow (n+1)!\geq nn!\geq aa^n=a^{n+1}[/tex] which yields that [tex]n+1\geq a\Rightarrow(n+1)!\geq a^{n+1}[/tex]

Therefore, [tex]n\geq a\Rightarrow n!\geq a^n[/tex]
But for [tex]n=3,a=2[/tex] using the inequality we just proved [tex]3\geq 2\Rightarrow3!\geq 2^3\Leftrightarrow 6\geq 8[/tex] Impossible!. Where is my mistake?

[EDIT] Don't bother answering. I have highlighted the mistake I made that rendered the inequality invalid for a greater than 1
 
Last edited:
Mathematics news on Phys.org


3.1415926535 said:
Suppose that [tex]n\geq a\Rightarrow n!\geq a^n, n\in \mathbb{N}-\left \{ 0 \right \}[/tex]

That's not the hypothesis used in induction.

Ordinary induction would use the hypothesis:
[itex]n!\geq a^n[/itex] where we think of [itex]n[/itex] as a particular integer, not as "all integers".

and you would have to prove [itex](n+1)! \geq a^{n+1}[/itex]

So-called "strong induction" would use the hypothesis:
For each integer [itex]i: 0 < i \leq n , i! \geq a^i[/itex]

and you would still have to prove [itex](n+1)! \geq a^{n+1}[/itex]
 


Stephen Tashi said:
That's not the hypothesis used in induction.

Ordinary induction would use the hypothesis:
[itex]n!\geq a^n[/itex] where we think of [itex]n[/itex] as a particular integer, not as "all integers".

and you would have to prove [itex](n+1)! \geq a^{n+1}[/itex]

So-called "strong induction" would use the hypothesis:
For each integer [itex]i: 0 < i \leq n , i! \geq a^i[/itex]

and you would still have to prove [itex](n+1)! \geq a^{n+1}[/itex]

I am sorry, I messed up with latex. That was not my hypothesis...
 


I was having trouble too. The world will probably end because of clerical errors.
 


The problem is that [itex]n+1 \geq a[/itex] does not imply that [itex]n \geq a[/itex] which you seem to have used.
 


The basic problem seems to be that allowed values of a increase with n, while induction would work only if a is constant.