论文信息 - On the convergence of quasi-random sampling/importance resampling

On the convergence of quasi-random sampling/importance resampling

This article discusses the general problem of generating representative point sets from a distribution known up to a multiplicative constant. The sampling/importance resampling (SIR) algorithm is known to be useful in this context. Moreover, the quasi-random sampling/importance resampling (QSIR) scheme, based on quasi-Monte Carlo methods, is a more recent modification of the SIR algorithm and was empirically shown to have better convergence. By making use of quasi-Monte Carlo theory, we derive upper bounds for the error of the QSIR scheme.

Ronald Cools | Bart Vandewoestyne

[1] J. Gentle. Random number generation and Monte Carlo methods , 1998 .

[2] D. Rubin,et al. The calculation of posterior distributions by data augmentation , 1987 .

[3] L. Devroye. Non-Uniform Random Variate Generation , 1986 .

[4] Giray Ökten,et al. Error reduction techniques in quasi-monte carlo integration , 1999 .

[5] M. J. Rufo,et al. Quasi-Random Sampling Importance Resampling , 2005 .

[6] Parris K. Egbert,et al. Importance resampling for global illumination , 2005, EGSR '05.

[7] D. Rubin. Using the SIR algorithm to simulate posterior distributions , 1988 .

[8] Alan E. Gelfand,et al. Bayesian statistics without tears: A sampling-resampling perspective , 1992 .

[9] K. Fang,et al. Number-theoretic methods in statistics , 1993 .

[10] I. Sobol. On the distribution of points in a cube and the approximate evaluation of integrals , 1967 .

[11] H. Niederreiter. Low-discrepancy and low-dispersion sequences , 1988 .

[12] J. Halton. On the efficiency of certain quasi-random sequences of points in evaluating multi-dimensional integrals , 1960 .

[13] H. Faure. Discrépance de suites associées à un système de numération (en dimension s) , 1982 .

[14] Jürgen Hartinger,et al. Quasi-Monte Carlo algorithms for unbounded, weighted integration problems , 2004, J. Complex..