home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!elroy.jpl.nasa.gov!swrinde!network.ucsd.edu!galaxy!guitar!baez
- From: baez@guitar.ucr.edu (john baez)
- Newsgroups: sci.math
- Subject: Approximation and Estimation (HELP!!)
- Message-ID: <24069@galaxy.ucr.edu>
- Date: 22 Nov 92 22:34:03 GMT
- Sender: news@galaxy.ucr.edu
- Organization: University of California, Riverside
- Lines: 60
- Nntp-Posting-Host: guitar.ucr.edu
-
- In article <1ejbebINNber@roundup.crhc.uiuc.edu> hougen@focus.csl.uiuc.edu (Darrell Roy Hougen) writes:
- >I need to estimate a 'hill' function between -pi/2 and pi/2. It must
- >be 0 at -pi/2 and pi/2 and it must be 1 at 0. It must increase
- >monotonically from 0 to 1 on [-pi/2,0] and decrease monotonically from
- >1 to 0 on [0,pi/2]. It must be smooth, ie., differentiable everywhere
- >in (-pi/2,pi/2), and have a derivative of 0 at 0.
- >
- >My first question is, does there exist a basis for functions of the
- >above form?
-
- Sure, infinitely many. It's best to say what *kind* of functions In
- other words, smooth, continuous, or L^2 functions. But in any of the case
- I just mentioned, there is a countable basis.
-
- >One possible basis that I've been thinking about is (cos(x))^k with k
- >ranging from -inf to +inf. All the members of this set satisfy the
- >conditions given above and it appears that any weighted sum of such
- >functions will also although it would be nice to prove that.
-
- Nope, it's not true, because all the functions you listed are "even"
- (satisfy f(x) = f(-x)), so you can't approximate anything but even
- functions. Also for k negative the functions you list are unbounded,
- which is technically a pain in the butt.
-
- The best basis to use is sin(2kx) and cos(2kx), taken together.
- This basis is used in the Fourier transform and is well-known to
- be a complete basis. I recommend this basis because a huge amount
- is known about it and there is a huge amount of software written to
- deal with it.
-
- >My second question is, if I have a potential basis, how can I prove
- >that it is a complete basis? In other words, how can I prove that any
- >f of the above form can be approximated arbitrarily well by a sum of
- >functions in the basis set?
-
- There are lots of ways. For L^2 functions
- it is sufficient for the functions to "separate
- points and span an algebra" to be a basis, IF they are linearly
- independent. I can explain these terms if it matters - they are
- properties that hold for the basis I suggested, but the "separates
- points" property doesn't hold for the basis you suggested.
-
- >My third question is, can the results be extended to two or three
- >dimensions? More specifically, assume that the function I am trying
- >to estimate, f(x,y,z), satisfies the above conditions in all three
- >dimensions. If necessary, assume additional smoothness conditions.
- >The same goes for the 1-D case above.
- >
- >In this case, I would like to use seperable functions as a basis. For
- >example, let g_ijk(x,y,z) = (cos(x))^i(cos(y))^j(cos(z))^k. Is this a
- >basis? Why or why not?
-
- If you have a basis f_i(x) of the smooth/continuous/L^2 functions
- on the interal (take your pick!), then one can form a basis
- f_i(x)f_j(y)f_k(z) for the same class of functions on the cube.
-
- Again, these are 3 theorems in analysis that I could prove for you,
- but I am too lazy right now. It takes a fair understanding of
- functional analysis to follow the proofs.
-
-