HARMMEAN calculates the harmonic mean of data elements. 
 The harmonic mean is the inverse of the mean of the inverse elements.
 
 	y = harmmean(x [,DIM [,W]]) is the same as 
 	y = mean(x,'H' [,DIM [,W]]) 

 DIM	dimension
	1 STD of columns
	2 STD of rows
	default or []: first DIMENSION, with more than 1 element
 W	weights to compute weighted mean (default: [])
	if W=[], all weights are 1. 
	number of elements in W must match size(x,DIM) 

 features:
 - can deal with NaN's (missing values)
 - weighting of data 
 - dimension argument also in Octave
 - compatible to Matlab and Octave

 see also: SUMSKIPNAN, MEAN, GEOMEAN

Package: nan