simulate_data: Generate data from a random DAG

In nicolagnecco/causalXtreme: Causal discovery in heavy-tailed models

Generate data from a random DAG

Description

Generate data from a random DAG.

Usage

simulate_data(
  n,
  p,
  prob_connect,
  distr = c("student_t", "gaussian", "log_normal"),
  tail_index = 1.5,
  has_confounder = FALSE,
  is_nonlinear = FALSE,
  has_uniform_margins = FALSE
)

Arguments

n	Positive integer. The number of observations, must be larger than 1.
p	Positive integer. The number of variables, must be larger than 1.
prob_connect	Numeric — between 0 and 1. The probability that an edge i {\rightarrow} j is added to the DAG.
distr	Character. The distribution of the noise. It is one of: student_t, in this case the user has to specify the tail_index, i.e., the degrees of freedom, gaussian, log_normal.
tail_index	Positive numeric. The tail index, i.e., degrees of freedom, of the noise.
has_confounder	Boolean. Are there confounders in the system?
is_nonlinear	Boolean. Is the data generated non linear?
has_uniform_margins	Boolean. Are the variables rescaled uniformly between 0 and 1?

Value

List. The list is made of:

• dataset — Numeric matrix. Dataset of simulated data with n rows and p columns (note that the hidden variables are not included in this matrix).

• dag — Square binary matrix. The generated DAG, including both the observed variables and the confounders, if has_confounder = TRUE.

• pos_confounders — Integer vector. Represents the position of confounders (rows and columns) in dag. If has_confounder = FALSE, then pos_confounders = integer(0).

nicolagnecco/causalXtreme documentation built on April 21, 2024, 4:22 a.m.