Register to reply

Prove that a normed space is not Banach

by russel
Tags: banach, normed, prove, space
Share this thread:
russel
#1
Feb16-12, 09:21 AM
P: 13
Hello everyone,
I have a problem and cannot solve it. Could you help? Here it is
We have a normed space and an uncountable Hamel basis of it. Prove that it is not a Banach space.
Should I use Baire theorem? Any suggestions?
Phys.Org News Partner Science news on Phys.org
Experts defend operational earthquake forecasting, counter critiques
EU urged to convert TV frequencies to mobile broadband
Sierra Nevada freshwater runoff could drop 26 percent by 2100
micromass
#2
Feb16-12, 09:50 AM
Mentor
micromass's Avatar
P: 18,333
You're not giving us much information to work with. If you want to show something not Banach, then try to find a Cauchy sequence which does not converge.
There are other ways to prove this of course. A lot depends on the space in question.
alan2
#3
Feb16-12, 10:37 AM
P: 206
You must have a particular space in mind that you didn't specify. An infinite dimensional Banach space necessarily has an uncountable Hamel basis so your result is not general. You need to show that your space is not complete. What is your space?

russel
#4
Feb16-12, 12:12 PM
P: 13
Prove that a normed space is not Banach

The exercise does not refer to a particular space. It is just a normed space X with an uncountable Hamel basis. A solution I came up with was to make a finite dimension closed subspace and show using baire that it is X, leading to a contadiction.
If the problem gave us a space for this example which one would that be?
micromass
#5
Feb16-12, 12:13 PM
Mentor
micromass's Avatar
P: 18,333
Could you quote the exercise exactly as it was given??
Bacle2
#6
Feb16-12, 12:39 PM
Sci Advisor
P: 1,170
Quote Quote by russel View Post
The exercise does not refer to a particular space. It is just a normed space X with an uncountable Hamel basis. A solution I came up with was to make a finite dimension closed subspace and show using baire that it is X, leading to a contadiction.
If the problem gave us a space for this example which one would that be?
But, how could you prove that a closed subspace of an undefined space equals the space itself?
micromass
#7
Feb16-12, 01:09 PM
Mentor
micromass's Avatar
P: 18,333
Also notice that a finite dimensional normed space is always Banach, so your proof is likely incorrect.


Register to reply

Related Discussions
Using Correlation :Random Variables as a Normed Space (Banach, Hilbert maybe.?) Set Theory, Logic, Probability, Statistics 3
Prove Finite Dimensional Normed Vector Space is Differentiable Calculus & Beyond Homework 2
Normed linear space and inner product space Calculus & Beyond Homework 8
Banach space Calculus & Beyond Homework 5
Banach space C(X) Differential Geometry 6