determineRank: Function to determine the cointegration rank using the rank...

In jonlachmann/sparsecoint: Forecasting using sparse cointegration

Function to determine the cointegration rank using the rank selection criterion

Description

Function to determine the cointegration rank using the rank selection criterion

Usage

determineRank(
  data,
  r.init = NULL,
  p,
  max.iter.lasso = 3,
  conv.lasso = 10^-2,
  max.iter.r = 5,
  beta.init,
  alpha.init,
  rho.glasso = 0.1,
  lambda.gamma,
  lambda_beta,
  intercept = FALSE,
  tol = 1e-04
)

Arguments

r.init	initial value of cointegration rank
p	number of lags to be included
max.iter.lasso	maximum number of iterations PML
conv.lasso	convergence parameter
max.iter.r	maximu number of iterations to compute cointegration rank
beta.init	initial value for beta
alpha.init	initial value for alpha
rho.glasso	tuning parameter for inverse covariance matrix
lambda.gamma	tuning parameter for GAMMA
tol	tolerance parameter glmnet function
Y	Response Time Series
X	Time Series in Differences
Z	Time Series in Levels
lambda.beta	tuning parameter for BETA

Value

A List containing; rhat: estimated cointegration rank it.r: number of iterations rhat_iterations: estimate of cointegration rank in each iteration mu: value of mu decomp: eigenvalue decomposition

jonlachmann/sparsecoint documentation built on April 14, 2022, 10:49 a.m.