Space of continuous functions C[a,b]

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
4 replies · 7K views
ninty
Messages
12
Reaction score
0
We know that dim(C[a,b]) is infinte. Indeed it cannot be finite since it contains the set of all polynomials.
Is the dimension of a Hamel basis for it countable or uncountable?

I guess if we put a norm on it to make a Banach space, we could use Baire's to imply uncountable.
I am however interested in a proof that does not rely on equipping the vector space with a norm.

If it's too long winded but available somewhere else(books, articles) kindly point it out to me and I'll read it.
 
Physics news on Phys.org
Hamel bases are always finite or uncountable.
 
JSuarez said:
Hamel bases are always finite or uncountable.

That's true if your space is an infinite dimensional normed vector space which is complete, and that result follows from Baire's category theorem AFAIK. So you haven't really strayed from the original proof.

A counting argument might suffice. Suppose there was a countable basis. There are only continuum many choices of finite linear combinations from these, but the cardinality of the vector space is larger than that
 
I used instead the fact that exponentials are continuous
Then E:={e^ax : a real } is a subset of C[a,b]

It's clear that for finite n, {e^ix : i in N} is linearly independent.
I'm fudging, but that would seem to imply that E is also linearly independent, since every finite subset is linearly independent.
Hence cardinality of a basis is at least |E| = continuum
 
Office_Shredder said:
A counting argument might suffice. Suppose there was a countable basis. There are only continuum many choices of finite linear combinations from these, but the cardinality of the vector space is larger than that

False: the cardinality of C[a,b] is the cardinality of the continuum. (Each function in C[a,b] is uniquely determined by its values on the rational numbers in [a,b].)

But yes, you could show that E = {eax | a real} is an uncountable linearly independent subset of C[a,b], to show that C[a,b] has no countable Hamel basis.

JSuarez said:
Hamel bases are always finite or uncountable.

The space of all real sequences which are eventually zero has a countable Hamel basis.
 
Last edited: