


 STATISTIC estimates various statistics at once.
 
 R = STATISTIC(x,DIM)
   calculates all statistic (see list of fun) in dimension DIM
   R is a struct with all statistics 
 y = STATISTIC(x,fun)
   estimate of fun on dimension DIM
   y gives the statistic of fun    
 DIM    dimension
    1: STATS of columns
    2: STATS of rows
     N: STATS of  N-th dimension 
    default or []: first DIMENSION, with more than 1 element
 fun    'mean'    mean
    'std'    standard deviation
    'var'    variance
    'sem'    standard error of the mean
    'rms'    root mean square
    'meansq' mean of squares
    'sum'    sum
    'sumsq'    sum of squares
    'CM%'    central moment of order %
    'skewness' skewness 
    'kurtosis' excess coefficient (Fisher kurtosis)
    'mad'    mean absolute deviation
 features:
 - can deal with NaN's (missing values)
 - dimension argument 
 - compatible to Matlab and Octave
 see also: SUMSKIPNAN
 REFERENCE(S):
 [1] http://www.itl.nist.gov/
 [2] http://mathworld.wolfram.com/