1

Here is my attempted proof:

$\forall n \in \mathbb{N}$, let $S_n$ be the statement: $\frac{2^{4n}-(-1)^n}{17} \in \mathbb{N}$

Base case: $S_1$: $\frac{2^{4(1)}-(-1)^1}{17} = \frac{16+1}{17} = 1 \in \mathbb{N}$

Inductive step: $\forall n \geq 1$, $S_n$ holds

$S_{n+1}$: $\frac{2^{4(n+1)}-(-1)^{n+1}}{17} = \frac{2^{4n} \cdot 2^4 +(-1)^{n}}{17} = \frac{2^{4n} \cdot (2^4+1) - 2^{4n} +(-1)^{n}}{17}$.

Now, we are left with $2^{4n} - \frac{2^{4n}-(-1)^n}{17}$.

We know $\frac{2^{4n}-(-1)^n}{17}$, $2^{4n} \in \mathbb{N}$ and $2^{4n} > \frac{2^{4n}-(-1)^n}{17} \implies 2^{4n} - \frac{2^{4n}-(-1)^n}{17} \in \mathbb{N}$

$$\tag*{$\blacksquare$}$$

Is my proof correct? I also feel there is a more elegant/smoother proof (assuming mine is correct).

  • 1
    Yes there is a more direct proof. Try to factor the numerator, $2^{4n} - (-1)^n$, as a difference of $n$th powers. – hardmath Oct 21 '18 at 16:23
  • See https://math.stackexchange.com/questions/2554949/prove-5n-2-cdot-3n-3-is-divisible-by-8-forall-n-in-mathbbn-using/2555077#2555077 – lab bhattacharjee Oct 21 '18 at 16:27

3 Answers3

1

Your proof is fine.

But note:

$(a - b)(\sum\limits_{k=0}^{n-1} a^kb^{n-1 -k}) =$

$a(\sum\limits_{k=0}^{n-1} a^kb^{n-1 -k}) - (\sum\limits_{k=0}^{n-1} a^kb^{n-1 -k})=$

$(\sum\limits_{k=0}^{n-1}a^{k+1}b^{n-1 - k}) -(\sum\limits_{k=0}^{n-1}a^kb^{n-k}) = $

$(\sum\limits_{j=1}^n a^jb^{n-j})-(\sum\limits_{j=0}^{n-1}a^jb^{n-j})=$

$([\sum\limits_{j=1}^{n-1} a^jb^{n-j}]+a^nb^{n-n}) - (a^0b^{n-0}+[\sum\limits_{j=1}^{n-1} a^jb^{n-j}]) = $

$a^n - b^n$.

So for all integers $a,b$ and $n\ge 1$ we will always have $a-b$ divides $a^n - b^n$

So $2^4 -(-1) = 17$ will always divide $(2^4)^n - (-1)^n$.

fleablood
  • 130,341
  • fleablood.Very clear, very nice. – Peter Szilas Oct 21 '18 at 17:38
  • fleablood.I delete it.Thanks:) – Peter Szilas Oct 21 '18 at 17:46
  • Simpler $ $ let $,x = a!-!b,$ in $,x\mid f(x)!=!(x!+!b)^n!-b^n,,$ by $,f(0) = 0\ \ $ QED $ $ Or, use congruencea, or the Binomial Theorem, etc. There are many simple ways to prove the Factor Theorem. – Bill Dubuque Oct 21 '18 at 18:08
  • Hmmm.... fair enough. But I think being direct is often best. ANd this is direct..... Any way $(a-b)|(a^n - b^n)$ is a basic result and no matter how you prove it, this question is an immediate result. – fleablood Oct 21 '18 at 22:33
0

Yes. Conceptually the induction follows very simply by multiplying the congruences below using CPR = Congruence Product Rule, $ $ i.e. raise $\ 16\equiv\, -1\ $ to the power $n$ by inductively multiplying

$$\begin{align}\bmod 17\!:\qquad \color{#c00}{16}\ &\equiv\ \color{#c00}{{-}1}\\ 16^{\large n}&\equiv (-1)^{\large n}\qquad\quad\ P(n)\\ \Rightarrow\ \ \color{#c00}{16}\,16^{\large n}&\equiv (-1)^{\large n}(\color{#c00}{-1})\quad\ P(n\!+\!1)\end{align}\quad\ \ \ $$

i.e. the proof is a special case of the (inductive) proof of the Congruence Power Rule.. Thus the use of congruences has highlighted innate arithmetical structure allowing us to reduce the induction to an easy one $\,a\equiv b\,\Rightarrow\, a^n\equiv b^n,\,$ with obvious inductive step: multiply by $\,a\equiv b\,$ via the product rule.

If you don't know congruences we can preserve this arithmetical essence by using an analogous product rule for divisibility, with $\ m\mid n\ $ meaning $\,m\,$ divides $\,n,\,$ namely

$$\qquad\ \begin {align} &17\mid\ \color{#c00}{16\,\ \ -\ \ B}\qquad\quad\ \ [B = -1\,\ {\rm in\ OP}]\\ &17\mid\ \ \ \ \ 16^{\large n} -\ B^{\large n}\qquad\ P(n)\\ \Rightarrow\ \ &17\mid\ \color{#c00}{16}16^{n}\! -\!\color{#c00}BB^{\large n}\qquad\ P(n\!+\!1) \end{align} $$

$\begin{align}{\bf Divisibility\ Product\ Rule}\ \ \ \ &m\mid\ a\ -\ b\qquad {\rm i.e.}\quad \ a\,\equiv\, b\\ &m\mid \ \ A\: -\: B\qquad\qquad \ A\,\equiv\, B\\ \Rightarrow\ \ &\color{}{m\mid aA - bB}\quad \Rightarrow\quad aA\equiv bB\!\pmod{\!m}\\[.2em] {\bf Proof}\,\ \ m\mid (\color{#0a0}{a\!-\!b})A + b(\color{#0a0}{A\!-\!B}) &\,=\, aA-bB\ \ \text{by $\,m\,$ divides $\rm\color{#0a0}{green}$ terms by hypothesis.}\end{align}$

You can find further discussion in many prior posts.

Bill Dubuque
  • 282,220
0

Alternatively, assuming $\frac{2^{4n}-(-1)^n}{17} \in \mathbb{N}$: $$\begin{align}\frac{2^{4(n+1)}-(-1)^{n+1}}{17} &= \frac{2^{4n} \cdot 2^4 +(-1)^{n}}{17} =\\ &=\frac{2^{4} \cdot (2^{4n}-(-1)^n) + 2^{4}(-1)^n +(-1)^{n}}{17}=\\ &=2^4\cdot \underbrace{\frac{2^{4n}-(-1)^n}{17}}_{\ge 1}+(-1)^{n}\in \mathbb N. \end{align}$$

farruhota
  • 32,168