Skip to contents

Generate sample data for LiM (3 mixed variables)

Source: R/generate_lim_sample.r
generate_lim_sample.Rd

Generates a small dataset with a known causal chain of continuous and discrete variables: x1 (continuous) -> x2 (discrete) -> x3 (continuous). Continuous variables use Laplace-distributed noise (non-Gaussian, as required by LiNGAM-family methods). By default the discrete variable is drawn from a Bernoulli distribution whose logit is a linear function of its parent; with is_poisson = TRUE it is instead drawn from a Poisson distribution whose log-mean is a linear function of its parent.

Usage

generate_lim_sample(n = 1000L, seed = NULL, is_poisson = FALSE)

Arguments

n

number of samples (default: 1000)

seed

random seed. If NULL (default), no seed is set and results are not reproducible across calls.

is_poisson

if TRUE, generate the discrete variable x2 as Poisson-distributed counts instead of binary 0/1 values (default: FALSE)

Value

A list with four elements:

  • data: data frame with columns x1, x2, x3.

  • adjacency_matrix: 3x3 true adjacency matrix, following the lingamr convention (m[to, from], i.e. adjacency_matrix["x2", "x1"] is the coefficient of the x1 -> x2 edge). The x1 -> x2 coefficient is on the linear-predictor scale (logit for binary, log for Poisson).

  • is_continuous: logical vector c(TRUE, FALSE, TRUE) marking which columns of data are continuous.

  • is_poisson: the input is_poisson flag, stored for reference.

Examples

dat <- generate_lim_sample(n = 500, seed = 1)
result <- lingam_lim(dat$data, is_continuous = dat$is_continuous)
result$adjacency_matrix
#>    x1 x2        x3
#> x1  0  1 0.1358125
#> x2  0  0 0.0000000
#> x3  0  1 0.0000000