m_step: Compute M step in the weighted ash problem for different...

In stephenslab/susiF.alpha: Sum of Single Functions

Compute M step in the weighted ash problem for different prior

Description

Compute M step in the weighted ash problem for different prior

Usage

m_step(
  L,
  zeta,
  indx_lst,
  init_pi0_w,
  control_mixsqp,
  nullweight,
  is.EBmvFR = FALSE,
  tol_null_prior = 0.001,
  ...
)

## S3 method for class 'lik_mixture_normal'
m_step(
  L,
  zeta,
  indx_lst,
  init_pi0_w,
  control_mixsqp,
  nullweight,
  is.EBmvFR = FALSE,
  tol_null_prior = 0.001,
  ...
)

## S3 method for class 'lik_mixture_normal_per_scale'
m_step(
  L,
  zeta,
  indx_lst,
  init_pi0_w = 1,
  control_mixsqp,
  nullweight,
  is.EBmvFR = FALSE,
  tol_null_prior = 0.001,
  ...
)

Arguments

L	output of L_mixsqp function
zeta	assignment probabilities for each covariate
indx_lst	list generated by gen_wavelet_indx for the given level of resolution, used only with class mixture_normal_per_scale
init_pi0_w	starting value of weight on null compoenent in mixsqp
control_mixsqp	list of parameter for mixsqp function see mixsqp package
nullweight	numeric value for penalizing likelihood at point mass 0 (should be between 0 and 1) (usefull in small sample size) @param tol_null_prior tolerance for the mixture on the null component if the mass on the point mass is large than 1- tol_null_prior then set BF=1
...	Other arguments.

Value

a vector of proportion (class pi_mixture_normal)

stephenslab/susiF.alpha documentation built on June 11, 2025, 1:09 p.m.