Skip to contents

Data simulation

Source: R/simulate.R
simulate.Rd

Simulates data with grouped predictor variables.

Usage

simulate(
  family = "gaussian",
  n0 = 100,
  n1 = 10000,
  n_group = 20,
  n_type = 2,
  size_group = c(5, 3),
  effect_size = c(1, 1),
  corfac_feature = 0.5,
  corfac_type = 0.5,
  corfac_group = 0.25,
  n_group_causal = 2,
  prop_causal = 0.5,
  noise_factor = 1,
  plot = FALSE,
  trial = FALSE
)

Arguments

family

character "gaussian", "binomial", "poisson", or "cox"

n0

number of training observations

n1

number of testing observations

n_group

number of variable groups

n_type

number of variable types

size_group

size of variable groups (per variable type): integer vector of length n_type

effect_size

effect sizes (per variable type): numeric vector of length n_type

corfac_feature

decrease of correlation if different variable: scalar in unit interval

corfac_type

decrease of correlation if different type: scalar in unit interval

corfac_group

decrease of correlation if different group: scalar in unit interval

n_group_causal

number of causal groups: integer

prop_causal

proportion of causal features within causal groups: scalar in unit interval

noise_factor

noise factor: numeric scalar

plot

Attempt to visualise effects of and correlation between variables? (TRUE or FALSE)

trial

logical (groups of negatively correlated subgroups)

Value

Returns a list with the following slots:

  • \(n_0 \times p\) matrix x_train

  • \(p\)-dimensional vector type

  • \(p\)-dimensional vector group

  • \(n_0\)-dimensional vector y_train

  • \(n_1 \times p\) matrix x_test

  • \(n_1\)-dimensional vector y_test

  • \(p\)-dimensional vector beta

  • data frame info with entries \(n_0\), \(n_1\), \(p\), n_type, n_group, and family

Examples

data <- corila:::simulate()
dims <- function(x) {
   if (is.matrix(x)||is.data.frame(x)) {
     paste(base::dim(x), collapse = " x ")
   } else {
     paste0(base::length(x))
   }
}
sapply(X = data, FUN = dims)
#>       x_train          type         group       y_train        x_test 
#>   "100 x 160"         "160"         "160"         "100" "10000 x 160" 
#>        y_test          beta          info 
#>       "10000"         "160"       "1 x 6"