-3

The question is to prove - $2^n>n^2$ (n is natural number >4) How do I prove it?

Soham
  • 9,990

3 Answers3

1

Prove it true for 5 and assume it true for some k.

So, $2^k>k^2$

Not prove it true for k+1.

We observe,$[2^{k+1}=2^k\cdot2]>[2\cdot k^2=k^2+k^2]$ (as assumed above)

Also, $2^{k+1}>k^2+3k$ (from the previous statement above as $k>4$ and so $k^2>3k$)

So, $2^{k+1}>k^2+2k+k$

So, $2^{k+1}>k^2+2k+1$ (as $k>1$)

So, $2^{k+1}>(k+1)^2$

Hence,proved

möbius
  • 2,443
Soham
  • 9,990
1

Alternatively, show by differentiation that the gradient of $f(x) = 2^x - x^2$ is always positive for $n \ge 4$. Hence conclude using the implication that $f(x)$ is strictly increasing (for $n \ge 4$), that

$$\begin{align}n > 4 &\implies f(n) > f(4) = 0\\&\implies 2^n - n^2 > 0\\&\implies 2^n > n^2\end{align}$$

Yiyuan Lee
  • 14,435
1

Of course induction is the best choice here, but if you want to use calculus, then you can discover a deeper result:

After squaring both sides, it's equivalent to proving $4^n>n^4$ for $n>4$ or equivalently $\sqrt[4]{4}>\sqrt[n]{n}$ for $n>4$. This is the famous result that the function $f(x)=x^{\frac{1}{x}}$ attains its minimum at $x=e$ and is strictly increasing for $x>e$. This is equivalent to $a^b>b^a$ for all $a,b$ (not necessarily integers!) with $a>b>e$

Tintarn
  • 2,931