computebic: Compute BIC scores for model selection

In seonjoo/smm: Sparse Mediation For High-Dimensional Mediators

Description

Compute BIC scores for model selection

Usage

computebic(
  X,
  Y,
  M,
  a,
  b,
  c,
  tol = 10^(-10),
  max.iter = 100,
  grpgroup = c(1, rep(1:ncol(M) + 1, 2)),
  threshold = 0
)

Arguments

X	One-dimensional predictor
Y	Outcome
M	Multivariate mediator
tol	(default -10^(-10)) convergence criterion
max.iter	(default=100) maximum iteration
grpgroup	(default=c(1,rep( 1:V +1,2)))
lambda	(default=log(1+(1:50)/125)) tuning parameter for L1 penalization
penalty.factor	(default=c(0,rep(sqrt(2),V))) give different weight of penalization for the 2V mediation paths.

Value

c directeffect

hatb Path b (M->Y given X) estimates

hata Path a (X->M) estimates

medest Mediation estimates (a*b)

alpha

lambda

nump Number of selected mediation paths

Author(s)

Seonjoo Lee, [e-mail]

References

TBA

Examples

N=100
V=50
set.seed(1234)
a = rbinom(V,1,0.1)*5;b<-a
X = rnorm(N)
M =  X %*% t(a)+ matrix(rnorm(N*V),N,V)
Y =  X + M %*% b + rnorm(N)

seonjoo/smm documentation built on Feb. 11, 2021, 5:54 a.m.