nts.beta: Function to estimate beta and omega First step: Determine...

In jonlachmann/sparsecoint: Forecasting using sparse cointegration

Function to estimate beta and omega First step: Determine Beta conditional on Gamma, alpha and Omega using Lasso Regression Second step: Determine Omega conditional on Gamma, alpha and beta using GLasso

Description

Function to estimate beta and omega First step: Determine Beta conditional on Gamma, alpha and Omega using Lasso Regression Second step: Determine Omega conditional on Gamma, alpha and beta using GLasso

Usage

nts.beta(
  Y,
  X,
  Z,
  gamma,
  rank,
  P,
  alpha,
  alphastar,
  lambda = NULL,
  rho_omega,
  cutoff,
  intercept = F,
  exo = NULL,
  tol = 1e-04
)

Arguments

Y	Response Time Series
X	Time Series in Differences
Z	Time Series in Levels
gamma	estimate of short-run effects
rank	cointegration rank
P	transformation matrix P derived from Omega
alpha	estimate of adjustment coefficients
alphastar	estimate of transformed adjustment coefficients
lambda	tuning paramter cointegrating vector
rho_omega	tuning parameter inverse error covariance matrix
cutoff	cutoff value time series cross-validation approach
intercept	F do not include intercept, T include intercept in estimation short-run effects
tol	tolerance parameter glmnet function
Omega	estimate of inverse error covariance matrix

Value

A list containing: BETA: estimate of cointegrating vectors OMEGA: estimate of inverse covariance matrix

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