i) if $ 0 < y < 1$ show $y^n \rightarrow 0$ when $n \rightarrow \infty$
ii) if $ y > 1 $ then $y^{\frac{1}{n}} \rightarrow 1$
I know it is not practice to ask two questions at once, but I figured since they are of similar ilk then it would be more helpful to combine them.
Attempt i)
I am already aware of one form of establishing this convergence by using the Monotone Convergence theorem. Our professor suggested another way to establish the convergence and I was trying to do it in that fashion.
Proof:
By the Archimedian property of the real numbers we know there exists a $p \in \mathbb{N}$ such that $0 < y < \frac{p}{p+1} < 1$. So it would be enough to show that: $$\Bigg(\frac{p}{p+1}\Bigg)^n \rightarrow 0$$
and then conclude by the squeeze theorem: $$y^n \rightarrow 0$$
I am trying to prove this using the definition of a limit of a sequence, so after expanding using the binomial theorem what I am trying to show is: $$\forall \epsilon >0, \ \exists \ N \in \mathbb{N} \ s.t.\ n \geq N \Rightarrow \Bigg| \frac{p^n}{p^n +np^{n-1} + \ldots + 1} - 0 \Bigg| < \epsilon$$
Notice: $p^n +np^{n-1} + \ldots + 1 > p^n + 1$..........
This is where I get stuck, I am trying to produce a N to satisfy my definition. I thought of possibly using logarithms but that was not fruitful. I am probably missing some simple trick, but alas I have not figured it out. What am I missing here?
Attempt ii)
For this question I want to use the Monotone convergence theorem to establish the result. The difficulty I am having is actually showing that $$y^{\frac{1}{n}} > y^{\frac{1}{n+1}}$$
Apparently by the ordered field axioms this already stands, but I don't recall establishing this result. If that is the case then It would be monotonically decreasing. But I don't think we've established this and I want to know how. What am I missing here?
Thank you for the help.
For i),
$$\Bigg| \frac{p^n}{p^n +np^{n-1} + \ldots + 1} - 0 \Bigg| < \Bigg| \frac{p^n}{p^n +np^{n-1}} \Bigg| = \Bigg| \frac{1}{1 +\frac{n}{p}} \Bigg|$$
For ii), since $y>0$, if not then $${(y^{\frac{1}{n}})}^{n(n+1)} \leq {(y^{\frac{1}{n+1}})}^{n(n+1)}$$ $$y^{n+1} \leq y^n$$ which is clearly impossible with $y>1$. Is this sufficient?
As an alternative way to do ii), write $y^{\frac{1}{n}}=1+\delta_n$. It suffices to show that $\delta_n \rightarrow 0$. By Bernoulli's inequality, $y=(1+\delta_n)^n \geq 1+n\delta_n$. For here $y$ serves as an upper bound the proof is finished.
This should be one of the standard ways for such limit.