tbackfitting: Perform backfitting starting at output of 'tgreedy'.

Description Usage Arguments Author(s)

Description

Perform backfitting starting at output of tgreedy.

Usage

1
2
3
4
5
tbackfitting(Y, factor_list, sigma_est, maxiter_bf = 100, tol_bf = 10^-6,
  maxiter_vem = 100, var_type = c("homoscedastic", "kronecker"),
  mixcompdist = "normal", alpha = 0, beta = 0, nullweight = 10,
  known_factors = NULL, known_modes = NULL, homo_modes = NULL,
  tol_r1 = 10^-3)

Arguments

Y

An array of numerics. The data.

factor_list

A list of matrices with the same number of columns. These are the starting values for the backfitting algorithm. The intended starting values are can be obtained from tgreedy.

sigma_est

Either a vector of estimated precisions (when var_type = "homoscedastic") or a list of matrices whose columns are estimated precisions (when var_type = "kronecker").

maxiter_bf

A positive integer. The maximum number of backfitting steps to perform.

tol_bf

A positive numeric. The stopping criterion for the backfitting algorithm.

maxiter_vem

A positive integer. The maximum number of steps in each VEM algorithm to perform at each iteration of the backfitting algorithm.

var_type

A string. What variance model should we assume? Options are homoscedastic noise ("homoscedastic") or Kronecker structured variance (kronecker).

mixcompdist

The mixing distribution to assume. Defaults to normal. Options are those available in the ashr package.

alpha

A non-negative numeric. The prior shape parameter for the variance. Defaults to zero.

beta

A non-negative numeric. The prior rate parameter for the variance. Defaults to zero.

nullweight

A numeric greater than or equal to 1. The penalty term on the probability of zero.

known_factors

A list of known factors for the modes indicated in known_modes. Defaults to NULL, where all factors are assumed to be unknown.

known_modes

A vector of integers. The modes that are known. Should be the same length as known_factors.

homo_modes

A vector of integers. If var_type = "kronecker" then homo_modes indicates which modes are assumed to be homoscedastic.

tol_r1

A positive numeric. The tolerance for the rank 1 runs of tflash.

Author(s)

David Gerard


kkdey/flashr documentation built on May 20, 2019, 10:36 a.m.