Skip to content

A procedure to reduce the parameters estimate in Granger-causal analysis through temporal smoothing.

Notifications You must be signed in to change notification settings

Eden-Kramer-Lab/spline-granger-causality

Repository files navigation

spline-granger-causality

A procedure to reduce the parameters estimate in Granger-causal analysis through temporal smoothing.

Code to infer networks using a modified version of conditional Granger causality as described in A procedure to increase power of Granger-causal analysis through temporal smoothing.

Must have MVGC Multivariate Granger Causality toolbox to run as implemented in:

Barnett, L., Seth, A.K., 2014. The MVGC multivariate Granger causality toolbox: A new approach to Granger-causal inference. J. Neurosci. Methods 223, 50–68. doi:10.1016/j.jneumeth.2013.10.018. .

Three example simulations are provided:

Example 1) main_sim_1N_high_freq.m Example 2) main_sim_1N_low_freq.m Example 3) main_sim_9N.m

Examples 1 and 2 fit models for one signal dominated by high or low frequencies, respectively. Example 3 fits networks to a multivariate system of nine signals.

To execute an example, call the simualtion at the MATLAB command line:

>> main_sim_1N_high_freq

About

A procedure to reduce the parameters estimate in Granger-causal analysis through temporal smoothing.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages