regrVonPS {kpcalg} | R Documentation |
Check if variable can be regressed to independence on its parents
Description
Uses the generalised additive model gam to non-linearly and non-parametrically regress variable V on its parents and set of variables S.
Usage
regrVonPS(G, V, S, suffStat, indepTest = kernelCItest, alpha = 0.2)
Arguments
G |
adjacency matrix, for the graph |
V |
integer, node which we regress |
S |
integer(s), set we regress on |
suffStat |
sufficient statistics to perform the independence test kernelCItest |
indepTest |
independence test to check for dependence between residuals of V and S |
alpha |
numeric cutoff for significance level of individual partial correlation tests |
Value
regrVonPS() returns the number of p-values smaller than the cutoff, i.e 0 means residuals of V are independent of all variables in S
Author(s)
Petras Verbyla (petras.verbyla@mrc-bsu.cam.ac.uk)
[Package kpcalg version 1.0.1 Index]