tinit_kron_components: Obtain initial estimates of each mode's components.

In kkdey/flashr: Factor Loading Adaptive SHrinkage in R

Description

The inital values of each component are taken to be proportional to the first singular vector of each mode-k matricization of the data array.

Usage

tinit_kron_components(Y, which_na = NULL, start = c("first_sv", "random"),
  known_factors = NULL, known_modes = NULL, homo_modes = NULL)

Arguments

Y	An array of numerics. The data.
which_na	Either NULL or an array the same dimension as Y indicating if an element of Y is missing (TRUE) or observed (FALSE).
start	How should we choose the starting values? Either using the first singular vector along each mode ("first_sv") or randomly ("random").
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.

Value

ex_list A list of vectors of the starting expected values of each component.

ex2_list A vector of starting values of E[x^2]

Author(s)

David Gerard

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