fciMI: Estimate a PAG by the FCI-MI Algorithm for Multiple Imputed...

In bips-hb/micd: Multiple Imputation in Causal Graph Discovery

Estimate a PAG by the FCI-MI Algorithm for Multiple Imputed Data Sets of Continuous Data

Description

This function is a modification of pcalg::fci() to be used for multiple imputation.

Usage

fciMI(
  data,
  alpha,
  labels,
  p,
  skel.method = c("stable", "original"),
  type = c("normal", "anytime", "adaptive"),
  fixedGaps = NULL,
  fixedEdges = NULL,
  NAdelete = TRUE,
  m.max = Inf,
  pdsep.max = Inf,
  rules = rep(TRUE, 10),
  doPdsep = TRUE,
  biCC = FALSE,
  conservative = FALSE,
  maj.rule = FALSE,
  verbose = FALSE
)

Arguments

data	An object of type mids, which stands for 'multiply imputed data set', typically created by a call to function mice()
alpha	Significance level (number in (0,1) for the conditional independence tests
labels	(Optional) character vector of variable (or "node") names. Typically preferred to specifying p.
p	(Optional) number of variables (or nodes). May be specified if labels are not, in which case labels is set to 1:p.
skel.method	Character string specifying method; the default, "stable" provides an order-independent skeleton, see pcalg::skeleton() for details.
type	Character string specifying the version of the FCI algorithm to be used. See pcalg::fci() for details.
fixedGaps	See pcalg::fci() for details.
fixedEdges	See pcalg::fci() for details.
NAdelete	See pcalg::fci() for details.
m.max	Maximum size of the conditioning sets that are considered in the conditional independence tests.
pdsep.max	See pcalg::fci() for details.
rules	Logical vector of length 10 indicating which rules should be used when directing edges. The order of the rules is taken from Zhang (2008).
doPdsep	See pcalg::fci() for details.
biCC	See pcalg::fci() for details.
conservative	See pcalg::fci() for details.
maj.rule	See pcalg::fci() for details.
verbose	If true, more detailed output is provided.

Value

See pcalg::fci() for details.

Author(s)

Original code by Diego Colombo, Markus Kalisch, and Joris Mooij. Modifications by Ronja Foraita.

Examples

daten <- windspeed[,1]
for(i in 2:ncol(windspeed)) daten <- c(daten, windspeed[,i])
daten[sample(1:length(daten), 260)] <- NA
daten <- matrix(daten, ncol = 6)

## Impute missing values
imp <- mice(daten, printFlag = FALSE)
fc.res <- fciMI(data = imp, label = colnames(imp$data), alpha = 0.01)

if (requireNamespace("Rgraphviz", quietly = TRUE))
plot(fc.res)

bips-hb/micd documentation built on Feb. 22, 2023, 7:59 p.m.