generateL: Generate a model with specified structure

In adallak/SCPackage: Learning Smooth Cholesky Factors of Large Covariance Matrices

Description

Generate a model with structure described below.

Usage

generateL(p, band = NULL, scaled = FALSE, case = c("a", "b", "c",
  "d"), seed = 25, scale = 1, ...)

Arguments

p	the dimension of L
band	Number of subdiagonals
case	admits for possible case a,b,c, and d discribed in Dallakyan and Pourahmadi(2019)
seed
scale	scales the subdiagonal by given constant to avoid large values.
...
stand	logical, If TRUE the diagonals of lower triangular matrix L are equal 0, otherwise log((1:p)/10+2)

Details

By default diagonal elements of L is of the form log((1:p)/10+2.

Value

Returns Standard Cholesky factor L and modified Cholesky Factor T.

Examples

generateL(p = 10, band = 4, case = "c")

adallak/SCPackage documentation built on Feb. 7, 2022, 1:58 a.m.