Quasi-regression

被引：61

作者：

An, J ^{[1
]}

Owen, A ^{[1
]}

机构：

[1] Stanford Univ, Stanford, CA 94305 USA

来源：

JOURNAL OF COMPLEXITY | 2001年 / 17卷 / 04期

关键词：

computer experiments; function mining; kriging; numerical noise; quasi-interpolation;

D O I：

10.1006/jcom.2001.0588

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

Quasi-regression is introduced for approximation of functions on the unit cube in s dimensions. It is computationally efficient, compared to kriging, for problems requiring a large number of function evaluations. This paper describes bow to implement quasi-regression and shows how to estimate the approximation error using the same data used to build the approximation. Four example functions are investigated numerically. (C) 2001 Academic Press.

引用

页码：588 / 607

页数：20

共 19 条

[1]

[Anonymous], [No title captured]

[2]

[Anonymous], STAT DECISION THEORY, DOI DOI 10.1007/978-1-4613-8768-8_20

[3]

Caflisch R. E., 1997, Journal of Computational Finance, V1, P27

[4] ALGORITHMS FOR COMPUTING THE SAMPLE VARIANCE - ANALYSIS AND RECOMMENDATIONS [J].