13
$\begingroup$

I'm looking for a reference. I am considering the situation where $X$ is a Banach space and $Y$ is a closed finite co-dimensional subspace. I am looking for a $W$ that is a complement of $Y$ (i.e. so that $X$ is the topological direct sum of $W$ and $Y$). I want them to be as far from parallel as possible.

The sense that I'm toying with (but I'm happy to have other senses proposed - I suspect they're all fairly equivalent) is that I want to consider the map $\Phi\colon Y\times W\to X$ given by $\Phi(y,w)=y+w$ to have the property that $\|\Phi^{-1}\|$ has small norm (say you put the norm $\|(y,w)\|=\max(\|y\|,\|w\|)$ on the product.

So I'm looking for $W$ that makes $\|\Phi^{-1}\|$ small. The question then is: how small can you make this quantity? I think I can make it less than $6^n$ if $Y$ is $n$-codimensional (irrespective of the space $X$), but I presume there are better bounds and references out there.

Improve this question
$\endgroup$

2 Answers 2

Reset to default
15
$\begingroup$

You can make the norm of $\|\Phi^{-1}\|$ to be of order $\sqrt n$. This is basically a theorem of Kadets and Snobar. A good reference is III.B.11 in Wojtaszczyk, P., Banach spaces for analysts, Cambridge Studies in Advanced Mathematics, 25. Cambridge University Press, Cambridge, 1991.

Improve this answer
$\endgroup$
0
0
$\begingroup$

I think these articles should meet their expectations, at least temporarily

http://www.springerlink.com/content/wk3121768n46462x/fulltext.pdf

http://arxiv.org/PS_cache/math/pdf/0501/0501048v1.pdf

Improve this answer
$\endgroup$
1
  • $\begingroup$ The link to springerlink.com is broken. I'm also unable to find any copy saved on the Wayback Machine. $\endgroup$
    – The Amplitwist
    Sep 18, 2022 at 13:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.