Description Usage Arguments Details Value Author(s) References Examples
View source: R/sparse.mediation.grplasso.R
Fit a mediation model via penalized maximum likelihood and structural equation model. The regularization path is computed using group lasso at a grid of values for the regularization parameter lambda. Currently, mediation analysis is developed based on gaussian assumption.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 |
X |
One-dimensional predictor |
M |
Multivariate mediator |
Y |
Outcome |
tol |
(default -10^(-10)) convergence criterion |
max.iter |
(default=100) maximum iteration |
lambda1 |
(default=exp(-5:0)) tuning parameter for c',a, b coefficients |
lambda2 |
(default=exp(seq(0,0.5*log(ncol(M)),length=3))) tuning parameter for the Omega=Sigma_2^-1 |
alpha |
(default=0) alpha=0, group lasso penalization will run |
group.penalty.factor |
(V+1)-dimensional group penalty factor vector. If a user does not want to penalize mediator, specify 0 otherwise 1. The first element is the direct effect followed by V-mediators. The default value is c(0,rep(1,V)). |
penalty.factor |
(1+2*V)-dimensional sparsity penalty factor vector. |
threshold |
(default=10^(-8)) |
non.zeros.stop |
(default=ncol(M)) when to stop searching regularization path |
Multiple Mediaton Model: (1) M = Xa + e1 (2) Y = Xc' + Mb + e2 And in the optimization, we do not regularize c', due to the assumption of partial mediation.
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
Seonjoo Lee, sl3670@cumc.columbia.edu
TBA
1 2 3 4 5 6 7 8 9 10 11 |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.