fOptions (version 3042.86)

LowDiscrepancy: Low Discrepancy Sequences

Description

A collection and description of functions to compute Halton's and Sobol's low discrepancy sequences, distributed in form of a uniform or normal distribution.

The functions are:

runif.halton	Uniform Halton sequence,
rnorm.halton	Normal Halton sequence,
runif.sobol	Uniform scrambled Sobol sequence,
rnorm.sobol	Normal scrambled Sobol sequence,
runif.pseudo	Uniform pseudo random numbers,

Usage

runif.halton(n, dimension, init) 
rnorm.halton(n, dimension, init)
runif.sobol(n, dimension, init, scrambling, seed) 
rnorm.sobol(n, dimension, init, scrambling, seed)
runif.pseudo(n, dimension, init)
rnorm.pseudo(n, dimension, init)

Arguments

dimension

an integer value, the dimension of the sequence. The maximum value for the Sobol generator is 1111.

init

a logical, if TRUE the sequence is initialized and restarts, otherwise not. By default TRUE.

n

an integer value, the number of random deviates.

scrambling

an integer value, if 1, 2 or 3 the sequence is scrambled otherwise not. If 1, Owen type type of scrambling is applied, if 2, Faure-Tezuka type of scrambling, is applied, and if 3, both Owen+Faure-Tezuka type of scrambling is applied. By default 0.

seed

an integer value, the random seed for initialization of the scrambling process. By default 4711. On effective if scrambling>0.

Value

All generators return a numeric matrix of size n by dimension.

Details

Halton's Low Discrepancy Sequences:

Calculates a matrix of uniform or normal deviated halton low discrepancy numbers.

Scrambled Sobol's Low Discrepancy Sequences:

Calculates a matrix of uniform and normal deviated Sobol low discrepancy numbers. Optional scrambling of the sequence can be selected.

Pseudo Random Number Sequence:

Calculates a matrix of uniform or normal distributed pseudo random numbers. This is a helpful function for comparing investigations obtained from a low discrepancy series with those from a pseudo random number.

References

Bratley P., Fox B.L. (1988); Algorithm 659: Implementing Sobol's Quasirandom Sequence Generator, ACM Transactions on Mathematical Software 14, 88--100.

Joe S., Kuo F.Y. (1998); Remark on Algorithm 659: Implementing Sobol's Quaisrandom Seqence Generator.

Examples

# NOT RUN {
## *.halton - 
   par(mfrow = c(2, 2), cex = 0.75)
   runif.halton(n = 10, dimension = 5)
   hist(runif.halton(n = 5000, dimension = 1), main = "Uniform Halton", 
     xlab = "x", col = "steelblue3", border = "white")
   rnorm.halton(n = 10, dimension = 5)
   hist(rnorm.halton(n = 5000, dimension = 1), main = "Normal Halton", 
     xlab = "x", col = "steelblue3", border = "white")
   
## *.sobol - 
   runif.sobol(n = 10, dimension = 5, scrambling = 3)
   hist(runif.sobol(5000, 1, scrambling = 2), main = "Uniform Sobol", 
     xlab = "x", col = "steelblue3", border = "white")
   rnorm.sobol(n = 10, dimension = 5, scrambling = 3)
   hist(rnorm.sobol(5000, 1, scrambling = 2), main = "Normal Sobol", 
     xlab = "x", col = "steelblue3", border = "white")
   
## *.pseudo - 
   runif.pseudo(n = 10, dimension = 5)
   rnorm.pseudo(n = 10, dimension = 5) 
# }