Quarterly of Applied Mathematics

Quarterly of Applied Mathematics

Online ISSN 1552-4485; Print ISSN 0033-569X



Asymptotic properties of best $ L_{2}[0,\,1]$ approximation by splines with variable knots

Authors: D. L. Barrow and P. W. Smith
Journal: Quart. Appl. Math. 36 (1978), 293-304
MSC: Primary 41A15
DOI: https://doi.org/10.1090/qam/508773
MathSciNet review: 508773
Full-text PDF Free Access

Abstract | References | Similar Articles | Additional Information

Abstract: Let $ S_N^k$ be the set of $ k$th-order splines on $ \left[ {0, 1} \right]$ having at most $ N - 1$ interior knots, counting multiplicities. We prove the following sharp asymptotic behavior of the error for the best $ {L_2}\left[ {0, 1} \right]$ approximation of a sufficiently smooth function $ f$ by the set $ S_N^k$:

$\displaystyle \mathop {\lim }\limits_{N \to \infty } {N^k}dist\left( {f, S_N^k}... ...ight)}}\left( \tau \right)} \right\vert} }^\sigma }d\tau } \right)^{1/\sigma }}$

, where $ \sigma = {\left( {k + 1/2} \right)^{ - 1}}$ and $ {B_{2k}}$ is the $ 2k$th Bernoulli number. Similar results have previously been obtained for piecewise polynomial (i.e., with no continuity constraints) approximation, but with different constant before the integral term. The approach we use is first to study the asymptotic behavior of dist($ f$, $ S_N^k\left( t \right)$), where $ S_N^k\left( t \right)$ is the linear space of $ k$th-order splines having simple knots determined from the fixed function $ t$ by the rule $ {t_i} \\ = t\left( {i/N} \right),i = 0,...,N$.

References [Enhancements On Off] (What's this?)

Similar Articles

Retrieve articles in Quarterly of Applied Mathematics with MSC: 41A15

Retrieve articles in all journals with MSC: 41A15

Additional Information

DOI: https://doi.org/10.1090/qam/508773
Article copyright: © Copyright 1978 American Mathematical Society

Brown University The Quarterly of Applied Mathematics
is distributed by the American Mathematical Society
for Brown University
Online ISSN 1552-4485; Print ISSN 0033-569X
© 2017 Brown University
Comments: qam-query@ams.org
AMS Website