VARDetect (version 0.1.8)

first.step.blocks.group: block fused sparse group lasso step (first step).

Description

Perform the block fused lasso to detect candidate break points.

Usage

first.step.blocks.group(
  data.temp,
  lambda.1.cv,
  lambda.2.cv,
  q,
  max.iteration = max.iteration,
  tol = tol,
  cv.index,
  blocks,
  group.case = "columnwise",
  group.index
)

Value

A list object, which contains the followings

brk.points

a set of selected break point after the first block fused lasso step

cv

the cross validation values for tuning parmeter selection

cv1.final

the selected lambda_1

cv2.final

the selected lambda_2

Arguments

data.temp

input data matrix, with each column representing the time series component

lambda.1.cv

tuning parmaeter lambda_1 for fused lasso

lambda.2.cv

tuning parmaeter lambda_2 for fused lasso

q

the AR order

max.iteration

max number of iteration for the fused lasso

tol

tolerance for the fused lasso

cv.index

the index of time points for cross-validation

blocks

the blocks

group.case

group sparse pattern: column, row.

group.index

group index for group sparse case