×

The construction of good extensible Korobov rules. (English) Zbl 1131.11051

Given the point set \(P_s=\{h\vec z/n +\Delta: h=1,\dots,n-1\} \subset [0,1)^s\), where \(z=(1,a,\dots,a^{s_i-1})\) for some \(a\in \{1,2,\ldots, n-1\}\) and \(\Delta \in [0,1)^s\) is chosen i.i.d. the authors want to find \(a\) such that the worst case error for QMC-integration is small. As the authors point out, for applications it is important to find lattice rules that give small QMC-integration errors simultaneously in several dimensions \(\mathcal S=\{s_1,\dots, s_d\}\). In this sense the algorithm given in the paper finds good \(a\)’s. In the last section the authors give also numerical examples to confirm their algorithm.

MSC:

11K45 Pseudo-random numbers; Monte Carlo methods
65C05 Monte Carlo methods
65D30 Numerical integration
Full Text: DOI

References:

[3] Dick, J.: The construction of extensible polynomial lattice rules with small weighted star discrepancy (submitted 2006). · Zbl 1135.11039
[4] Dick, J., Pillichshammer, F., Waterhouse, B. J.: The construction of good extensible lattices (submitted 2006). · Zbl 1211.11092
[7] Hickernell, F. J.: Lattice rules: how well do they measure up? In: Random and quasi-random point sets (Hellekalek, P. and Larcher, G., eds.). Lecture Notes in Statistics, vol. 138. New York: Springer, pp. 109–166 (1998). · Zbl 0920.65010
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.