New GraphVar version: beta 0.5 online!
DOWNLOAD HERE
Updates:
1. Implementation of sparse inverse covariance estimation (SICE) with the graphical lasso:
- Generate covariance matrices from time courses that should be used with the SICE threshold function (network construction)
- Network construction: will use SICE to produce binary matrices with selected target densities using the input covariance matrices
2. Implementation of a new time-series randomizer:
- Multivariate algorithm from Prichard, D., & Theiler, J. (1994). Generating surrogate data for time series with several simultaneously measured variables.
Physical Review Letters, 73(7), 951.
- Basically works like this: will cause randomizing the observed time series by taking its Fourier transform, scrambling its phase and then inverting the transform.