qmle_ruv2_lambda_grid: Quasi-mle when first 'vr' rows of 'Y' have variances which...

In dcgerard/vicar: Various Ideas for Confounder Adjustment in Regression

Description

This one will iterate through a grid of lambdas and choose the one with the highest likelihood.

Usage

qmle_ruv2_lambda_grid(
  Y,
  r,
  vr,
  gridsize = 20,
  gridlow = 1,
  gridhigh = 10,
  tol = 10^-3,
  maxit = 1000,
  plot_llike = FALSE
)

Arguments

Y	A matrix of numerics. The data.
r	the rank.
vr	The number of the first few rows whose variances differ by a multiplicative factor.
gridsize	A positive integer. The size of the grid for the variance inflation parameter.
gridlow	A positive numeric. The minimum variance inflation parameter to try out.
gridhigh	A positive numeric. The maximum variance inflation parameter to try out.
tol	The function tolerance for stopping. A positive numeric.
maxit	The maximum number of iterations to run.
plot_llike	A logical. Should I plot the profiled log-likelihood?

Value

alpha A matrix of the estimated factor loadings.

Z A matrix of the estimated factors.

sig_diag A vector of the estimated column-wise variances.

lambda A numeric scalar of the estimated variance inflation parameter.

Author(s)

David Gerard

dcgerard/vicar documentation built on July 7, 2021, 1:08 p.m.