estimate_var_lasso: Estimate VAR-lasso model on multivariate time series

In bdemeshev/torro:

Description

estimate_var_lasso returns an estimated VAR-lasso model

Usage

estimate_var_lasso(y, h = 1, p = 12, struct = "OwnOther", gran = c(25,
  10), T1 = floor(nrow(as.matrix(y))/3), T2 = floor(2 *
  nrow(as.matrix(y))/3))

Arguments

y	multivariate time series
h	forecasting horizon
p	number of lags
struct	type of VAR-lasso
gran	granularity vector, If gran = (a, b) then cross-validation checks b values form lambda to lambda/a.
T1	left border for cross validation samples
T2	right border for cross validation samples

Details

Just a wrapper for cv.BigVAR function from BigVAR package. This function does NOT normalise variables. Pre-normalisation is required by the idea.

Value

estimated VAR-lasso model

Examples

data(rus_macro)
y_small <- rus_macro[, c("cpi", "employment", "m2")]
y_small_scaled <- scale_to(y_small)
var_lasso_model <- estimate_var_lasso(y_small_scaled)

bdemeshev/torro documentation built on May 21, 2019, 8:36 a.m.