解析函数的稳定外推
Stable extrapolation of analytic functions
AI总结:
本文研究了在[-1,1]等距网格上给出扰动样本的情况下,如何稳定地外推x>1的解析函数。通过构造最优外推多项式近似,证明了在满足过采样条件时,外推误差随扰动减小而收敛,且具有x依赖的分数幂衰减率。
摘要:
本文探讨了在x>1范围内解析函数的外推问题,给定来自[-1,1]等距网格的扰动样本。数学 folklore 表明外推通常病态严重,但本文展示更精确的陈述具有有趣 nuances。对于在Bernstein 椭圆内解析的函数f,以及函数样本的均匀扰动水平ε,我们构造了渐近最优外推多项式e(x)作为显式给出的最小二乘多项式近似。证明e(x)在区间I_ρ∈[1,(ρ+ρ^{-1})/2)内,当ε→0时,以x依赖的分数幂速率收敛于f(x)。更精确地说,对于每个x∈I_ρ,|f(x)-e(x)|=O(ε^{-log r(x)/logρ}),其中r(x)=(x+√(x²-1))/ρ,前提是满足过采样条件。即M*≤(1/2)√N,这是近似理论所必需的。简而言之,外推在特征光滑长度的分数范围内具有弱稳定性。函数样本数量N+1不影响外推误差大小,只要满足过采样条件。此外,本文还证明了无法从N+1个等距样本构造比e(x)更准确的渐近外推多项式,无论使用线性还是非线性方法。证明涉及关于等距样本在Chebyshev基下多项式近似的稳定性原陈述,这些结果具有独立价值。
英文摘要:
This paper examines the problem of extrapolation of an analytic function for $x > 1$ given perturbed samples from an equally spaced grid on $[-1,1]$. Mathematical folklore states that extrapolation is in general hopelessly ill-conditioned, but we show that a more precise statement carries an interesting nuance. For a function $f$ on $[-1,1]$ that is analytic in a Bernstein ellipse with parameter $ρ> 1$, and for a uniform perturbation level $ε$ on the function samples, we construct an asymptotically best extrapolant $e(x)$ as a least squares polynomial approximant of degree $M^*$ given explicitly. We show that the extrapolant $e(x)$ converges to $f(x)$ pointwise in the interval $I_ρ\in[1,(ρ+ρ^{-1})/2)$ as $ε\to 0$, at a rate given by a $x$-dependent fractional power of $ε$. More precisely, for each $x \in I_ρ$ we have \[ |f(x) - e(x)| = \mathcal{O}\left( ε^{-\log r(x) / \logρ} \right), \qquad\qquad r(x) = \frac{x+\sqrt{x^2-1}}ρ, \] up to log factors, provided that the oversampling conditioning is satisfied. That is, \[ M^* \leq \frac{1}{2} \sqrt{N}, \] which is known to be needed from approximation theory. In short, extrapolation enjoys a weak form of stability, up to a fraction of the characteristic smoothness length. The number of function samples, $N+1$, does not bear on the size of the extrapolation error provided that it obeys the oversampling condition. We also show that one cannot construct an asymptotically more accurate extrapolant from $N+1$ equally spaced samples than $e(x)$, using any other linear or nonlinear procedure. The proofs involve original statements on the stability of polynomial approximation in the Chebyshev basis from equally spaced samples and these are expected to be of independent interest.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。