arguments_var_lasso_toy: Arguments for VAR-lasso models toy version

In bdemeshev/torro:

Description

Tibble with the following columns

Usage

data(arguments_var_lasso_toy)

Format

tibble

Details

• p number of lags

• struct LASSO lag structure Three top players according to https://arxiv.org/pdf/1508.07497.pdf page 23 are SparseLag (Lag Sparse Group VARX-L), OwnOther (Own/Other Group VARX-L), SparseOO (Own/Other Sparse Group VARX-L). but SparseLag and SparseOO hangs the computer.

• gran_1 first granularity parameter: left grid search border is equal to maximal lambda / gran_1. Maximal labmda sets all coefficient estimates to zero.

• gran_2 second granularity parameter: number of grid points for cross-validation. Values gran_1 = 25 and grand_2 = 10 are suggested by BigVAR user's guide, http://www.wbnicholson.com/BigVAR.pdf, page 9.

• pars_id id of parameter combination

• expand_window_by number of observations added to basic window length. Usually this number is equal to the number of parameters in the model. Three for an average ARIMA model. Can not be computed beforehand for auto.arima because the number of parameters is not known a-priori.

See also full version, arguments_var_lasso.

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