rBAP: Causal discovery with latent confounding and non-Gaussian...

In ysamwang/ngBap: Estimates BAP structures assuming non-Gaussian errors

Causal discovery with latent confounding and non-Gaussian data

Description

Samples from a linear SEM corresponding to a BAP

Usage

rBAP(
  n,
  p,
  dist,
  d,
  b,
  B = NULL,
  Omega = NULL,
  directLow = 0.6,
  directHigh = 1,
  bidirectLow = 0.3,
  bidirectHigh = 0.5,
  t.df = 13,
  ancestral = F,
  shuffle = T,
  signs = T
)

Arguments

n	number of observations
p	number of variables
dist	the distribution of the errors "gauss" for normal distribution "t" for T distribution with t.df degrees of freedom "gamma" for gamma distribution "unif" for uniform distribution "laplace" for Laplace distribution "lognormal" for lognormal distribution
d	number of directed edges
b	number of bidirected edges
B	the direct effects; leave as null to randomly sample
Omega	the covariance of epsilon; leave as null to randomly sample
directLow	the lower bound for the absolute value of a randomly sampled directed edge weight
directHigh	the upper bound for the absolute value of a randomly sampled directed edge weight
bidirectLow	the lower bound for the absolute value of a randomly sampled bidirected edge weight
bidirectHigh	the upper bound for the absolute value of a randomly sampled bidirected edge weight
t.df	degree of freedom if distribution is T
ancestral	T restricts the graph to be ancestral; F allows for a BAP which may not be ancestral
shuffle	Setting shuffle = T randomly relabels the nodes; if set to F then 1... p will always is a valid causal ordering
signs	Setting signs = F, all parameters are positive; setting signs = F allows both positive and negative values

Value

• B the directed edge coefficients

• Omega covariance of varepsilon

• directEdges the adjacency matrix of the directed edges (support of B)

• bidirectEdgesthe adjacency matrix of the bidirected edges (support of Omega)

• errs the sampled epsilon values

• Y the sampled Y values

• Sigma the population covariance

• ord a topological ordering of the nodes

ysamwang/ngBap documentation built on Nov. 1, 2022, 3:20 a.m.