groupMeans

mu = groupMeans( X, DIM, CATEG )

MEAN of the matrix X along dimension specified, but grouping the other dimensions according to the categories in 'categ'.

CATEG must threfore have same length as the specified dimension of X.

returns: a matrix with same size as X, except for the specified dimension -- along that dimension, there is one slice per unique level of 'categ'.

% example:
% X = [1 2 3; 4 5 6; 7 8 9; 10 11 12]
% C = [ 1; 1; 2; 2 ]; % each row of X has a category
% groupMeans( X, 1 ,C ); % mean along vertical dimension, but group
% % rows according to C
% ans =
% 2.5 3.5 4.5
% 8.5 9.5 10.5
%
% i.e. the first row is mean( X( C==1, : ) , 1 )
% and the second row is mean( X( C==2, : ) , 1 )
%

note: the function will use NANMEAN if you have it on the path.