Constructing Sobol Sequences with Better Two-Dimensional Projections
Abstract
Direction numbers for generating Sobol$'$ sequences that satisfy the so-called Property A in up to 1111 dimensions have previously been given in Joe and Kuo [ACM Trans. Math. Software, 29 (2003), pp. 49–57]. However, these Sobol$'$ sequences may have poor two-dimensional projections. Here we provide a new set of direction numbers alleviating this problem. These are obtained by treating Sobol$'$ sequences in d dimensions as $(t,d)$-sequences and then optimizing the t-values of the two-dimensional projections. Our target dimension is 21201.
[1] , An economic method of computing LP$_\tau$-sequences, Zh. Vychisl. Mat. Mat. Fiz., 19 (1979), pp. 243–245 (in Russian); Comput. Maths. Math. Phys., 19 (1980), pp. 252–256 (in English).
[2] , Algorithm $659$: Implementing Sobol's quasirandom sequence generator, ACM Trans. Math. Software, 14 (1988), pp. 88–100. ACMSCU 0098-3500
[3]
[4]
[5] , Constructing embedded lattice rules for multivariate integration, SIAM J. Sci. Comput., 28 (2006), pp. 2162–2188. SJOCE3 1064-8275
[6] , The construction of good extensible rank-1 lattices, Math. Comp., 77 (2008), pp. 2345–2373. MCMPAF 0025-5718
[7] , Pricing discrete dynamic fund protections, N. Am. Actuar. J., 7 (2003), pp. 23–31.
[8]
[9] , Remark on Algorithm $659$: Implementing Sobol's quasirandom sequence generator, ACM Trans. Math. Software, 29 (2003), pp. 49–57. ACMSCU 0098-3500
[10] , Component-by-component constructions achieve the optimal rate of convergence for multivariate integration in weighted Korobov and Sobolev spaces, J. Complexity, 19 (2003), pp. 301–320. JOCOEH 0885-064X
[11]
[12] , Modelling genetic networks with noisy and varied experimental data: The circadian clock in Arabidopsis thaliana, J. Theor. Biol., 234 (2005), pp. 383–393. JTBIAP 0022-5193
[13] , Quasi-random sequences and their discrepancies, SIAM J. Sci. Comput., 15 (1994), pp. 1251–1279. SJOCE3 1064-8275
[14]
[15] , Fast algorithms for component-by-component construction of rank-1 lattice rules in shift-invariant reproducing kernel Hilbert spaces, Math. Comp., 75 (2006), pp. 903–920. MCMPAF 0025-5718
[16] , Faster valuation of financial derivatives, J. Portfolio Manage., 22 (1995), pp. 113–120.
[17] , Projections of digital nets and sequences, Math. Comput. Simulation, 55 (2001), pp. 239–247. MCSIDR 0378-4754
[18] , Quasi-Monte Carlo in finance: Extending for problems of high effective dimension, Econ. Apl., 9 (2005), pp. 577–594.
[19]
[20] , Constructing randomly shifted lattice rules in weighted Sobolev spaces, SIAM J. Numer. Anal., 40 (2002), pp. 1650–1665. SJNAAM 0036-1429
[21] , Finite-order weights imply tractability of multivariate integration, J. Complexity, 20 (2004), pp. 46–74. JOCOEH 0885-064X
[22] , When are quasi-Monte Carlo algorithms efficient for high dimensional integrals?, J. Complexity, 14 (1998), pp. 1–33. JOCOEH 0885-064X
[23] , Distribution of points in a cube and approximate evaluation of integrals, Zh. V<#0233>chisl. Mat. Mat. Fiz., 7 (1967), pp. 784–802 (in Russian); Comput. Maths. Math. Phys., 7 (1967), pp. 86–112 (in English).
[24] , Uniformly distributed sequences with an additional uniform property, Zh. V<#0233>chisl. Mat. Mat. Fiz., 16 (1976), pp. 1332–1337 (in Russian); Comput. Maths. Math. Phys., 16 (1976), pp. 236–242 (in English).
[25] , The effective dimension and quasi-Monte Carlo integration, J. Complexity, 19 (2003), pp. 101–124. JOCOEH 0885-064X
[26] , Why are high-dimensional finance problems often of low effective dimension?, SIAM J. Sci. Comput., 27 (2005), pp. 159–183. SJOCE3 1064-8275
[27] , Low discrepancy sequences in high dimensions: How well are their projections distributed?, J. Comput. Appl. Math., 213 (2008), pp. 366–386. JCAMDI 0377-0427
