HDGC_VAR_multiple: Test multiple combinations Granger causality in High...

In Marga8/HDGCvar: Granger Causality Testing in High Dimensional Vector Autoregressive Models

Test multiple combinations Granger causality in High Dimensional mixed Integrated and Cointegrated VARs

Description

This function is a wrapper around HDGC_VAR that allows for multiple combinations to be tested

Usage

HDGC_VAR_multiple(
  data,
  GCpairs,
  p = 1,
  d = 0,
  bound = 0.5 * nrow(data),
  parallel = FALSE,
  n_cores = NULL
)

Arguments

data	the data matrix or object that can be coerced to a matrix.
GCpairs	it should contain a nested list. The outer list is all the pairs to be considered. The inner list contains the GCto and GCfrom vectors needed for HDGC_VAR.
p	lag length of the VAR
d	order of lag augmentation corresponding to suspected max order of integration
bound	lower bound on tuning parameter lambda
parallel	TRUE for parallel computing
n_cores	nr of cores to use in parallel computing, default is all but one

Value

LM Chi-square test statistics (asymptotic), LM F-stat with finite sample correction, both with their corresponding p-value. Lasso selections are also printed to the console.

References

Hecq, A., Margaritella, L., Smeekes, S., "Inference in Non Stationary High Dimensional VARs" (2020, check the latest version at https://sites.google.com/view/luca-margaritella )

Hecq, A., Margaritella, L., Smeekes, S., "Granger Causality Testing in High-Dimensional VARs: a Post-Double-Selection Procedure." arXiv preprint arXiv:1902.10991 (2019).

Examples

## Not run: GC=list(list("GCto"="Var 1","GCfrom"="Var 2"),list("GCto"="Var 2","GCfrom"="Var 3"))
## Not run: HDGC_VAR_multiple(sample_dataset_I1,GCpairs=GC,3,2)

Marga8/HDGCvar documentation built on May 25, 2024, 11:12 a.m.