Cauchy Convergence in Normed Vector Spaces

Adblax
Messages
7
Reaction score
0

Homework Statement



Fix a<b in R, and consider the two norms Norm(f)1:=Integralab( Modulus(f)) and Norm(f)Infinity:= sup{Mod(f(x)): a <= x <= b} on the vector space C[a,b]. This question shows that they are not equivalent.
a. Show that there is K in R such that for all f in C[a,b],
Norm(f)1 <= K*Norm(f)Infinity
b. Show that there is no K in R such that for all f in C[a,b],
Norm(f)infinity <= K*Norm(f)1
c. Consider the sequence of functions (fn) in C[-1,1] given by fn(x):=x1/(2n-1)
Is (fn) a cauchy sequence in (C[-1,1], Norm(f)1)? Why?
Is (fn) a cauchy sequence in (C[-1,1], Norm(f)Infinity)? Why?

Homework Equations



Cauchy Convergence Definition:
For all E>0, There exists M(E)>0, in the natural numbers, such that for all m>=n>=M(E), Norm(fm-fn)<E

The Attempt at a Solution



Found the first two parts relatively easy I think, for a taking k to be (b-a) and for b, considering the sequence of functions
fn(x):= { n2-nx for 0<=x<=1/n
0 otherwise
However, for the proof of cauchy sequences I am at a loss, as we cannot use normal convergence in the normed vector space as it converged to something that isn't continuous. So I am trying to use the definition of cauchy convergence, for the first part I have tried splitting up the integral into [0,1] and [-1,0] but have no idea for the second part.
 
Last edited:
Physics news on Phys.org
For the 1-norm this is a Cauchy sequence.
You need to prove that \|f_p-f_q\|_1\rightarrow 0as p,q\rightarrow +\infty.
So, you'll need to calculate

\|f_p-f_q\|_1=\int_{-1}^1{|x^{1/(2p-1)}-x^{1/(2q-1)}|}

and then prove that this expression goes to 0 as p,q\rightarrow +\infty.
 
I somewhat understand the first part yes, however I really cannot see what to do for the second part, as I know it doesn't have cauchy convergence, however, intuitively, the supremum tends to 0 eventually, as given an x, they will both tend to 1 as p,q tends to infinity, and thus the supremum tends to 0 (assuming x isn't 0)
 
C([-1,1]) with the sup-norm is complete, thus all Cauchy convergent sequences converges. But if a sequences converges uniformly, then it converges pointswize. So you need to show that the pointswise limit of your sequence is not continuous.
 
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...
Back
Top