Skip to contents

Mirrors the common workflow:

  1. tetris(..., fixed_bigT=FALSE)

  2. big_T <- choose.A(fit, ...)

  3. tetris(..., fixed_bigT=TRUE, A_fixed=big_T)

Usage

fit_tetris_2step(
  Y_list,
  alpha = "auto",
  beta = 1,
  nrun = 3000,
  burn = 2000,
  nprint = 200,
  chooseA_args = list(),
  ...
)

Arguments

Y_list

List of study matrices.

alpha

Either numeric alpha, or "auto" to use ceiling(1.25 * S).

beta

Beta hyperparameter.

nrun, burn, nprint

MCMC controls for both runs.

chooseA_args

List of extra arguments passed to choose.A().

...

Extra args passed to tetris() in both runs.

Value

A bifa_fit (fixed refit) with stage1 and big_T attached.