#9 global fit

[sebastian_busch]

It would be great to have an option to say something
like: A1 is the same for all Q, fit this value. Or W1 =
D*Q. fit D.

[sebastian_busch]

Logged In: YES
user_id=1528569

a workaround might be to take the mean of the fitted values.
as this is a quite tedious work to do by hand, one can list
the values of the fitted parameter and then copy them one by
one as arguments of the following script which simply
computes the mean of the arguments.

#!/bin/bash

mean=0.0
zaehler=0
for i in $@
do
mean=$(echo "scale=9; $mean+$i" | bc)
let "zaehler+=1"
done
mean=$(echo "scale=2; $mean/$zaehler" | bc)

echo $mean

[Florian Kargl]

Logged In: YES
user_id=1481723
Originator: NO

follow up to the suggestions by sebastian_busch.

Mean values of curve fitted values can easily be
calculated within FRIDA by using

1 ci
'number of parameter'

If 1 is the fitted curve this results in a second file (2)
containing the parameter versus the z-value (spectra denominator
e. g. Q for S(Q,w)).

The mean value can then be calculated by summation
2 oi
15
*
3

The result has to be divided by the number of
points.

Another possible solution is to fit a
linear curve with slope 0 to the data.