This shows you the differences between two versions of the page.

Link to this comparison view

reference:ft_connectivity_mutualinformation [2018/08/23 14:43]
reference:ft_connectivity_mutualinformation [2018/08/23 14:43] (current)
Line 1: Line 1:
 +Note that this reference documentation is identical to the help that is displayed in MATLAB when you type "help ft_connectivity_mutualinformation"​.
 +  <a href=/​reference/​ft_connectivity_mutualinformation><​font color=green>​FT_CONNECTIVITY_MUTUALINFORMATION</​font></​a>​ computes mutual information using the information
 +  breakdown toolbox (ibtb), as described in Magri et al., BMC Neuroscience 2009,
 +  1471-2202.
 +  Use as
 +    mi = ft_connectivity_mutualinformation(data,​ ...)
 +  The input data should be a Nchan x Nobservations matrix.
 +  Additional optional input arguments come as key-value pairs:
 +    histmethod = The way that histograms are generated from the data. Possible values
 +                 are '​eqpop'​ (default), '​eqspace',​ '​ceqspace',​ '​gseqspace'​.
 +                 See the help of the '​binr'​ function in the ibtb toolbox for more information.
 +    numbin ​    = scalar value. The number of bins used to create the histograms needed for
 +                 the entropy computations
 +    opts       = structure that is passed on to the '​information'​ function in the ibtb
 +                 ​toolbox. See the help of that function for more information.
 +    refindx ​   = scalar value or '​all'​. The channel that is used as '​reference channel'​.
 +    refdata ​   = 1xNobservations vector, as an alternative to the refindx. Refdata takes precedence over refindx
 +  The output contains the estimated mutual information between all channels and the reference channel(s).
 +  See also <a href=/​reference/​ft_connectivityanalysis><​font color=green>​FT_CONNECTIVITYANALYSIS</​font></​a>​