Prove the lower bound for a sequence (Buck, Advanced Calculus)

Click For Summary

Homework Help Overview

The discussion revolves around proving a lower bound for a sequence defined recursively, specifically in the context of advanced calculus. Participants explore various approaches to establish the relationship between terms in the sequence and their growth behavior.

Discussion Character

  • Exploratory, Mathematical reasoning, Problem interpretation

Approaches and Questions Raised

  • Participants suggest using induction to prove the lower bound, with some focusing on the relationship between consecutive terms. Others discuss the implications of specific inequalities and the sum of square roots. There are questions about the validity of certain expressions and the setup of the recursive definition.

Discussion Status

The discussion is active, with multiple participants contributing different perspectives and methods. Some have offered insights into the inductive approach, while others are questioning the assumptions and definitions used in the problem. There is no explicit consensus, but several productive lines of reasoning are being explored.

Contextual Notes

Some participants note the importance of base cases in induction and the need to clarify the definitions of terms in the sequence. There are also references to specific inequalities that may hold under certain conditions, indicating a need for careful consideration of the problem's constraints.

yucheng
Messages
232
Reaction score
57
Homework Statement
Define a sequence ##{x_n}## by $$x_1=1, x_{n+1} = x_n + \sqrt{x_n}$$

Show that ##\forall n\geq 1: x_n \geq \frac{n^2}{9}##

(Buck, Advanced Calculus, section 1.6, Exercise 16)
Relevant Equations
N/A
Clearly, ##x_{n+1}>x_n \because x_n + \sqrt{x_n} > x_n##
$$
\begin{align*}
x_{n+1} &= x_n+ \sqrt{x_n} \\
&= x_1 + \sqrt{x_1} + \sqrt{x_2} + \cdots \sqrt{x_n} \\
&>n+1
\end{align*}
$$
##\because \sqrt{x_n}>\sqrt{x_1}=1##

In fact, $$x_{n+1} > 1+ \sqrt{1} + \sqrt{2}+ \sqrt{3} + \cdots \sqrt{n}$$

So... Might this help? Closed form for sum of square roots
 
Physics news on Phys.org
Say y_n=x_n−x_{n−1} for n>1, y1=1 the relation is written as
y_{n+1}^2=x_n=\sum_{i=1}^n y_i
For n=1 the statement stands,i.e.
x_1=1>\frac{1^2}{9}
If it stand for n
y_{n+1}^2=x_n > \frac{n^2}{9}
y_{n+1} > \frac{n}{3}
Now you can show it stands for n+1.
 
Last edited:
  • Like
Likes   Reactions: yucheng
anuttarasammyak said:
Say y_n=x_n−x_{n−1} for n>1, y1=1 the relation is written as
y_{n+1}^2=x_n=\sum_{i=1}^n y_i
For n=1 the statement stands,i.e.
x_1=1>\frac{1^2}{9}
If it stand for n
y_{n+1}^2=x_n > \frac{n^2}{9}
y_{n+1} > \frac{n}{3}
You can show it stand for n+1.

did you mean this (I changed all to ##\#\###)<br /> <br /> Say ##y_n=x_n−x_{n−1}## for n&gt;1, y1=1 the relation is written as<br /> ##y_{n+1}^2=\sum_{i=1}^n y_i ##<br /> For n=1 the statement stands,i.e.<br /> ##x_1=1&gt;\frac{1^2}{9}##<br /> If it stand for n<br /> ##y_{n+1}^2=\sum_{i=1}^n y_i &gt; \frac{n^2}{9}##<br /> ##y_{n+1} &gt; \frac{n}{3}##<br /> You can show it stand for ##n+1##.
 
I think you can just do this by mostly straightforward induction. In particular notice that ##\frac{(n+1)^2}{9} < \frac{n^2}{9} + \sqrt{\frac{n^2}{9}},## at least if n is big enough (proof of this left to the reader)
 
  • Like
Likes   Reactions: yucheng, suremarc and pasmith
@Office_Shredder
I guess we always forget the most straightforward and basic methods :doh:thanks!
 
yucheng said:
@Office_Shredder
I guess we always forget the most straightforward and basic methods :doh:thanks!
So, did you find the answer?Tell us about it if you want.
 
  • Like
Likes   Reactions: epenguin
anuttarasammyak said:
Say y_n=x_n−x_{n−1} for n>1, y1=1 the relation is written as
y_{n+1}^2=x_n=\sum_{i=1}^n y_i
For n=1 the statement stands,i.e.
x_1=1&gt;\frac{1^2}{9}
If it stand for n
y_{n+1}^2=x_n &gt; \frac{n^2}{9}
y_{n+1} &gt; \frac{n}{3}
Now you can show it stands for n+1.
This is an different approach that i took to solve it. Does the equation ##x_n=\sum_{i=1}^n y_i ## hold? How did you show it? Because if you try ##\sum_{i=1}^n y_i=y_1+y_2+...+y_n=(x_1-x_0)+(x_2-x_1)+(x_3-x_2)+(x_4-x_3)+(x_5-x_4)+## ##...+(x_n-x_{n-1})+(x_{n+1}-x_n)=-x_0+x_{n+1}=-1+x_{n+1}##
 
##y_1=1## to be add and the final term be n not n+1.
 
anuttarasammyak said:
##y_1=1## to be add and the final term be n not n+1.
You are correct, I saw it now. Sorry for that.
 
  • #10
Office_Shredder said:
I think you can just do this by mostly straightforward induction. In particular notice that ##\frac{(n+1)^2}{9} < \frac{n^2}{9} + \sqrt{\frac{n^2}{9}},## at least if n is big enough (proof of this left to the reader)

infinitely small said:
So, did you find the answer?Tell us about it if you want.

Yup. Base case: ##x_1=1 \geq 1/9## Assume inductively that ##x_n \geq \frac{x^2}{9}##, then ##\frac{(n+1)^2}{9} < \frac{n^2}{9} + \sqrt{\frac{n^2}{9}} \leq x_n + \sqrt{x_n} = x_{n+1}##

P.S. ##\frac{3n}{9} > \frac{2n+1}{9} \because n>1## (except base case)
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
Replies
10
Views
2K
  • · Replies 25 ·
Replies
25
Views
4K
Replies
9
Views
2K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 24 ·
Replies
24
Views
3K
  • · Replies 13 ·
Replies
13
Views
3K
  • · Replies 9 ·
Replies
9
Views
3K
  • · Replies 17 ·
Replies
17
Views
2K