Parameters      Defaults

FSUM windows       
     /
     DISPLAY       
     CONDITIONS    
     LOOP          

     GRAPHIC       
     NET           
     TOTAL         
     FWHM          

     CONFIDENCE(co) 0.683
     STORE(st)     

Purpose

  Compute sum and moments.

Function

  Compute sum and moments of raw (input) or net (input minus fit) data.
  Output results on terminal, current picture or global parameter.

Example

  FSUM a,b,c / NET FWHM STORE( p(3,1) )

  Compute the moments of the net data (2nd moment expressed as FWHM)
  in the three display windows A,B,C.
  Results are stored in global parameter array p().
  p(3,1), p(3,2), ... receive moments of window A,
  p(4,1), p(4,2), ... receive moments of window B, and
  p(5,1), p(5,2), ... receive moments of window C.

List of windows, interpreted as follows:
window item = number: condition,
window item = letter: display window,
window item =      #: temporary window
window item =      *: all fit windows.

All display windows are integrated

All conditions are integrated

Windows are prompted in a loop. See also command FWINDOWS

Write results into current picture

Sum and moments of the net data are calculated, i.e. the values of the fit function
are subtracted from the input data.

Sum and moments of the input data are calculated.

The variance (second moment) is converted into an FWHM value according to
FWHM = sqrt( 8*log(2) * n / (n-1) ) * variance,
where n is the number of data points.
The skewness (third moment) is expressed in terms of the square root of the variance.

For 0 < co < 1 the evaluated function errors correspond to confidence limit co.
Usually co = 0.683 corresponds to the propagated error bars of the fitted input data.

Name of a floating point global parameter array to receive the calculated moments.
Storing starts at the specified index, or, if the index is omitted,
at the first element. If the array is 1-dimensional only sums are returned,
otherwise the data are stored columnwise, containing the following items
per row (sufficient bounds provided):
x-limits, sum, error, mean, variance (or FWHM), skewness.