update_B_L2: Updating loading matrix B with conave L2 norm penalty

In YipengUva/RpESCA: R implementaion of pESCA models with various concave penalties

Description

This is an intermediate step of the algorithm for fitting pESCA model. The details of this function can be found in ref thesis.

Usage

update_B_L2(JHk, A, B0, Sigmas0, d, fun_concave, alphas, rhos, lambdas,
  gamma)

Arguments

JHk	An output of the majorizaiotn step
A	The score matrix A during k-th iteration
B0	The loading matrix B during the previous iteration
Sigmas0	The group length during the previous iteration
d	A numeric vector contains the numbers of variables in different data sets
fun_concave	A string indicates the used concave function
alphas	The dispersion parameters of exponential dispersion families
rhos	An output of the majorizaiotn step
lambdas	A numeric vector indicates the values of tuning parameters for each data set.
gamma	The hyper-parameter of the concave penalty

Value

This function returns the updated loading matrix B.

Examples

## Not run: 
B <- update_B_L2(JHk,A,B0,Sigmas0,d,
                   fun_concave,alphas,rhos,lambdas,gamma)

## End(Not run)

YipengUva/RpESCA documentation built on July 2, 2019, 6:41 p.m.