A Python implementation of the VBA_groupBMC function for Bayesian model selection for group studies from the VBA-toolbox by Lionel Rigoux.
See docstrings in groupBMC.py for implementation details
and demo.ipynb for an example of usage.
The package can now be also installed using pip:
pip install groupBMC==1.0
from groupBMC.groupBMC import GroupBMC
import numpy as np
L = np.array([[-100, -120, -100, -120], [-120, -100, -120, -100], [-110, -110, -110, -110]])
result = GroupBMC(L).get_result()
References:
[1] Rigoux, L., Stephan, K. E., Friston, K. J., & Daunizeau, J. (2014). Bayesian model selection for group studies—revisited. NeuroImage, 84, 971-985. https://www.tnu.ethz.ch/fileadmin/user_upload/documents/Publications/2014/2014_Rigoux_Stephan_Friston_Daunizeau.pdf.
[2] Stephan, K. E., Penny, W. D., Daunizeau, J., Moran, R. J., & Friston, K. J. (2009). Bayesian model selection for group studies. NeuroImage, 46(4), 1004-1017. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2703732/pdf/ukmss-5226.pdf.