var_to_tsdata_nonstat

Generate random multi-trial non-stationary Gaussian VAR time series

Syntax

   X = var_to_tsdata_nonstat(A,SIG,N)

Arguments

See also Common variable names and data structures.

input

   A          time-varying VAR coefficients matrix
   SIG        time-varying residuals covariance matrix
   N          number of trials (default: 1)

output

   X          multi-trial Gaussian VAR time series
   E          residuals time series
   mtrunc     number of initial time steps truncated

Description

Return N independent random Gaussian non-stationary VAR time series, with time-varying coefficients A and residuals covariances SIG. The last index in A and SIG is the time index (the number of observations is taken as the size of the last dimension). If SIG is 2 dimensional, then it is replicated at each time.

Useful for generating test data; see mvgc_demo_nonstationary.

References

[1] L. Barnett and A. K. Seth, The MVGC Multivariate Granger Causality Toolbox: A New Approach to Granger-causal Inference, J. Neurosci. Methods 223, 2014 [ preprint ].

See also

var_to_tsdata | genvar_nonstat | mvgc_demo_nonstationary