selectVariables {valse} | R Documentation |
selectVariables
Description
For a given lambda, construct the sets of relevant variables for each cluster.
Usage
selectVariables(
phiInit,
rhoInit,
piInit,
gamInit,
mini,
maxi,
gamma,
glambda,
X,
Y,
thresh = 1e-08,
eps,
ncores = 3,
fast
)
Arguments
phiInit |
an initial estimator for phi (size: p*m*k) |
rhoInit |
an initial estimator for rho (size: m*m*k) |
piInit |
an initial estimator for pi (size : k) |
gamInit |
an initial estimator for gamma |
mini |
minimum number of iterations in EM algorithm |
maxi |
maximum number of iterations in EM algorithm |
gamma |
power in the penalty |
glambda |
grid of regularization parameters |
X |
matrix of regressors |
Y |
matrix of responses |
thresh |
real, threshold to say a variable is relevant, by default = 1e-8 |
eps |
threshold to say that EM algorithm has converged |
ncores |
Number or cores for parallel execution (1 to disable) |
fast |
boolean to enable or not the C function call |
Value
a list, varying lambda in a grid, with selected (the indices of variables that are selected), Rho (the covariance parameter, reparametrized), Pi (the proportion parameter)