MHB Proving (n/2)^n > n > (n/3)^n for all n > 6

  • Thread starter Thread starter juantheron
  • Start date Start date
AI Thread Summary
The discussion centers on proving the inequality (n/2)^n > n! > (n/3)^n for all natural numbers n > 6. The user demonstrates this for n = 600 using the values A = (300)^{600}, B = 600!, and C = (200)^{600}, concluding that A > B > C. The conversation highlights the use of Stirling's approximation and logarithmic properties to establish the inequalities. Participants suggest that the proof can be simplified by considering logarithms and reference the Maclaurin series for further validation. The thread concludes with a focus on the mathematical foundations of the inequalities and the methods to prove them without induction.
juantheron
Messages
243
Reaction score
1
If $A = (300)^{600}$ and $B=600!$ and $C = (200)^{600}$, Then arrangement of $A,B,C$ in Decreasing order, is

My approach:: Here $A>C$ and Using the formula $\displaystyle \left(\frac{n}{2}\right)^n>n! >\left(\frac{n}{3}\right)^n$ for all natural no. $n>6$

and put $n=600,$ we get $(300)^{600}>600!>(200)^{600}$ So $A>B>C$

My Question is How can we prove $\displaystyle \left(\frac{n}{2}\right)^n>n! >\left(\frac{n}{3}\right)^n$ for all natural no. $n>6$

Can we prove it without induction.

Thanks
 
Physics news on Phys.org
jacks said:
If $A = (300)^{600}$ and $B=600!$ and $C = (200)^{600}$, Then arrangement of $A,B,C$ in Decreasing order, is

My approach:: Here $A>C$ and Using the formula $\displaystyle \left(\frac{n}{2}\right)^n>n! >\left(\frac{n}{3}\right)^n$ for all natural no. $n>6$

and put $n=600,$ we get $(300)^{600}>600!>(200)^{600}$ So $A>B>C$

My Question is How can we prove $\displaystyle \left(\frac{n}{2}\right)^n>n! >\left(\frac{n}{3}\right)^n$ for all natural no. $n>6$

Can we prove it without induction.

Thanks
This should follow from Stirling's formula together with the fact that $2<e<3$.
 
jacks said:
If $A = (300)^{600}$ and $B=600!$ and $C = (200)^{600}$, Then arrangement of $A,B,C$ in Decreasing order, is

My approach:: Here $A>C$ and Using the formula $\displaystyle \left(\frac{n}{2}\right)^n>n! >\left(\frac{n}{3}\right)^n$ for all natural no. $n>6$

and put $n=600,$ we get $(300)^{600}>600!>(200)^{600}$ So $A>B>C$

My Question is How can we prove $\displaystyle \left(\frac{n}{2}\right)^n>n! >\left(\frac{n}{3}\right)^n$ for all natural no. $n>6$

Can we prove it without induction.

Thanks

The proof is made easier if You consider logarithms, because...

$\displaystyle \ln\ (\frac{n}{2})^{n} = n\ (\ln n - \ln 2)\ (1)$

$\displaystyle \ln\ (\frac{n}{3})^{n} = n\ (\ln n - \ln 3)\ (2)$

$\displaystyle \ln n! \sim n\ (\ln n - 1)\ (3)$

Take into account that $\displaystyle \ln 2 < 1 < \ln 3$...

Kind regards

$\chi$ $\sigma$
 
Thanks chisigma

would you like to explain me How we get $\ln(n!)\approx n\left(\ln (n)-1\right)$
 
jacks said:
Thanks chisigma

would you like to explain me How we get $\ln(n!)\approx n\left(\ln (n)-1\right)$

It consists in the Stirling's approximation that in more precise form is...

$\displaystyle \ln n! = n\ \ln n - n + \mathcal {O} (\ln n)\ (1)$ Stirling's Approximation -- from Wolfram MathWorld

Kind regards

$\chi$ $\sigma$
 
The lower bound $n!>\left(\frac{n}{3}\right)^n$ is easy to obtain from the Maclaurin series for $e^x$. We have $e^x=\sum_{n=0}^\infty \frac{x^n}{n!}$. In particular, $e^x>\frac{x^n}{n!}$. Taking $x=n$ gives $e^n>\frac{n^n}{n!}$, i.e., $n!>\left(\frac{n}{e}\right)^n> \left(\frac{n}{3}\right)^n$ since $e<3$.
 
jacks said:
Thanks chisigma

would you like to explain me How we get $\ln(n!)\approx n\left(\ln (n)-1\right)$

Alternatively... in...

http://mathhelpboards.com/discrete-mathematics-set-theory-logic-15/difference-equation-tutorial-draft-part-i-426.html#post2494

...it has been demonstrated that for x 'large enough' is... $\displaystyle \phi(x) = \frac{d}{d x} \ln x! \sim \ln x\ (1)$... so that... $\displaystyle \ln n! \sim \int_{0}^{n} \ln x\ dx = n\ \ln n - n\ (2)$ Kind regards$\chi$ $\sigma$
 
Back
Top