lags_upbound_BIC: Lag length Selection via BIC empirical upper bound

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

Lag length Selection via BIC empirical upper bound

Description

Selects the lag length p of the VAR using an empirical upper bound: residuals of the diagonalized VAR are used to build the empirical covariance matrix and an approximation of its determinant that uses the matrix trace is employed to be able to select p using Bayesian Information Criterion

Usage

lags_upbound_BIC(data, p_max = 10)

Arguments

data	a dataframe or matrix of the original set of time series forming the VAR
p_max	maximum lag length to consider, default is 10

Value

returns the estimated lag length upper bound

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

 lags_upbound_BIC(sample_dataset_I1, p_max=10)

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