IntMutualInfFromMargProb

 [MI]=IntMutualInfFromMargProb(NumbVar,Card,UnivProb,BivProb)
 IntMutualInfFromMargProb:   This functions computes the normalized matrix of mutual information (MI)
                             from the sets of univariate and bivariate marginal probabilities
 INPUTS:
 NumbVar: Number of variables
 Card: Vector with the dimension of all the variables. 
 UnivProb: Univariate probabilities
 BivProb: Bivariate probalities
 OUTPUTS: 
 MI: Matrix of mutual information 
 
 Last version 8/26/2008. Roberto Santana (roberto.santana@ehu.es)

0001 function[MI]=IntMutualInfFromMargProb(NumbVar,Card,UnivProb,BivProb)
0002 % [MI]=IntMutualInfFromMargProb(NumbVar,Card,UnivProb,BivProb)
0003 % IntMutualInfFromMargProb:   This functions computes the normalized matrix of mutual information (MI)
0004 %                             from the sets of univariate and bivariate marginal probabilities
0005 % INPUTS:
0006 % NumbVar: Number of variables
0007 % Card: Vector with the dimension of all the variables.
0008 % UnivProb: Univariate probabilities
0009 % BivProb: Bivariate probalities
0010 % OUTPUTS:
0011 % MI: Matrix of mutual information
0012 %
0013 % Last version 8/26/2008. Roberto Santana (roberto.santana@ehu.es)
0014 
0015 MI = zeros(NumbVar);
0016 for i=1:NumbVar-1,
0017  for j=i+1:NumbVar,
0018    for k=0:Card(i)-1,
0019       for l=0:Card(j)-1,
0020         if(BivProb{i,j}(Card(j)*k+l+1)>0)
0021             MI(i,j) = MI(i,j) + BivProb{i,j}(Card(j)*k+l+1)* log(BivProb{i,j}(Card(j)*k+l+1)/(UnivProb{i}(k+1)*UnivProb{j}(l+1)));
0022         end
0023       end,
0024    end,
0025    MI(i,j) = MI(i,j)/(Card(i)*Card(j)); %Normalization of the mutual information
0026    MI(j,i) = MI(i,j);
0027  end,
0028 end,