rebmix (version 2.8.0)

RNGMIX-methods: Random Univariate or Multivariate Finite Mixture Generation

Description

Returns as default the RNGMIX univariate or multivariate random datasets for mixtures of conditionally independent normal, lognormal, Weibull, gamma, binomial, Poisson or Dirac component densities. If model equals "RNGMVNORM" multivariate random datasets for mixtures of multivariate normal component densities with unrestricted variance-covariance matrices are returned.

Usage

## S3 method for class 'RNGMIX':
RNGMIX(model = "RNGMIX", Dataset.name = character(), 
       rseed = -1, n = numeric(), Theta = list(), ...)
## ... and for other signatures

Arguments

model

see Methods section below.

Dataset.name

a character vector containing list names of data frames of size $n \times d$ that d-dimensional datasets are written in.

rseed

set the random seed to any negative integer value to initialize the sequence. The first file in Dataset.name corresponds to it. For each next file the random seed is decremented $r_{\mathrm{seed}} = r_{\mathrm{seed}} - 1$. The default value

n

a vector containing numbers of observations in classes $n_{l}$, where number of observations $n = \sum_{l = 1}^{c} n_{l}$.

Theta

a list containing $c$ parametric family types pdfi. One of "normal", "lognormal", "Weibull", "gamma", "binomial", "Poisson" or "Dirac". Component para

...

currently not used.

Value

• Returns an object of class RNGMIX or RNGMVNORM.

Details

RNGMIX is based on the "Minimal" random number generator ran1 of Park and Miller with the Bays-Durham shuffle and added safeguards that returns a uniform random deviate between 0.0 and 1.0 (exclusive of the endpoint values).

References

W. H. Press, S. A. Teukolsky, W. T. Vetterling, and B. P. Flannery. Numerical Recipes in C: The Art of Scientific Computing. Cambridge University Press, Cambridge, 1992.

Examples

# Generate and print multivariate normal datasets with diagonal 
# variance-covariance matrices.

n <- c(75, 100, 125, 150, 175)

Theta <- list(pdf1 = rep("normal", 4),
  theta1.1 = c(10, 12, 10, 12),
  theta2.1 = c(1, 1, 1, 1),
  pdf2 = rep("normal", 4),
  theta1.2 = c(8.5, 10.5, 8.5, 10.5),
  theta2.2 = c(1, 1, 1, 1),
  pdf3 = rep("normal", 4),
  theta1.3 = c(12, 14, 12, 14),
  theta2.3 = c(1, 1, 1, 1),
  pdf4 = rep("normal", 4),
  theta1.4 = c(13, 15, 7, 9),
  theta2.4 = c(2, 2, 2, 2),
  pdf5 = rep("normal", 4),
  theta1.5 = c(7, 9, 13, 15),
  theta2.5 = c(3, 3, 3, 3))

simulated <- RNGMIX(Dataset.name = paste("simulated_", 1:25, sep = ""),
  rseed = -1,
  n = n,
  Theta = Theta)
  
simulated

# Generate and print multivariate normal datasets with unrestricted 
# variance-covariance matrices.

n <- c(200, 50, 50)

Theta <- list(pdf1 = rep("normal", 3),
  theta1.1 = c(0, 0, 0),
  theta2.1 = c(9, 0, 0, 0, 4, 0, 0, 0, 1),
  pdf2 = rep("normal", 3),
  theta1.2 = c(-6, 3, 6),
  theta2.2 = c(4, -3.2, -0.2, -3.2, 4, 0, -0.2, 0, 1),
  pdf3 = rep("normal", 3),
  theta1.3 = c(6, 6, 4),
  theta2.3 = c(4, 3.2, 2.8, 3.2, 4, 2.4, 2.8, 2.4, 2))

simulated <- RNGMIX(model = "RNGMVNORM",
  Dataset.name = paste("simulated_", 1:2, sep = ""),
  rseed = -1,
  n = n,
  Theta = Theta)
  
simulated

# Generate and print multivariate mixed continuous-discrete dataset.

n <- c(400, 100, 500)

Theta <- list(pdf1 = c("lognormal", "Poisson", "binomial", "Weibull"),
  theta1.1 = c(1, 2, 10, 2), 
  theta2.1 = c(0.3, NA, 0.9, 3),
  pdf2 = c("lognormal", "Poisson", "binomial", "Weibull"),
  theta1.2 = c(3.5, 10, 10, 10),
  theta2.2 = c(0.2, NA, 0.1, 7),
  pdf3 = c("lognormal", "Poisson", "binomial", "Weibull"),
  theta1.3 = c(2.5, 15, 10, 25), 
  theta2.3 = c(0.4, NA, 0.7, 20))
  
simulated <- RNGMIX(Dataset.name = paste("simulated_", 1:5, sep = ""),
  rseed = -1,
  n = n,
  Theta = Theta)
  
simulated  

# Generate and print univariate mixed Weibull dataset.

n <- c(75, 100, 125, 150, 175)

Theta <- list(pdf1 = "Weibull",
  theta1.1 = 12,
  theta2.1 = 2,
  pdf2 = "Weibull",
  theta1.2 = 10,
  theta2.2 = 4.1,  
  pdf3 = "Weibull",
  theta1.3 = 14,
  theta2.3 = 3.2,  
  pdf4 = "Weibull",
  theta1.4 = 15,
  theta2.4 = 7.1,  
  pdf5 = "Weibull",
  theta1.5 = 9,
  theta2.5 = 5.3)  

simulated <- RNGMIX(Dataset.name = "simulated",
  rseed = -1,
  n = n,
  Theta = Theta)
  
simulated