[AIT logo]

Institut für Astronomie und Astrophysik

Abteilung Astronomie

Sand 1, D-72076 Tübingen, Germany
[Uni logo]

fiterror Source code in cafe_fiterror.pro


           reconstruct error column by fitting model
           fiterror [,subgroup][,group][,/selected][,/quiet]
Input Parameters

           quiet      - Do not show fit processing.
           selected   - Apply fit to selected data points only.
           All options may be set with the "set" command. The
           command prefix is "fiterror".
          If the data set does not contain a error column the fit
          process assumes an error of 1 which leads to improper fit
          statistics. With this command it is possible to reconstruct
          an error using following assumptions:
          1.) The error is basically the same for all data points.
          2.) The model to fit is a good fit so the resulting reduced
              Chi^2 is 1.
          Of course these assumptions are not always justified, so
          this command should be used with care. This also means that
          for following fits the resulting reduced Chi^2 value is 1
          and could not be used to establish a goodness of the fit.
          But the command may help for deriving a formal error by
          first selecting ranges of data points with well-known model
          features, reconstructing the error column and then apply
          this full data set to other regions where a closer model
          must be applied.
          If more than one data set is given all data sets which are
          involved in the fitting process are taken into
          account. Excluded are:
          - Groups without a defined model
          - Groups with a model but no valid data points. If a group
            contains defined but not selected data points and the
            option /selected was chosen this group is also ignored.
          Subgroups with unselected data points when the group is
          valid and the select option is used are noticed.
          Use with care! Former error columns are modifierd; and the
          new error column of data may contain misleading values.
Side Effects
           Adds error column to data set.
           Changes parameter values/errors in environment
           according fit result.
           > model, lin      ; apply simple model for first part
           > ignore, 100-*   ; of data, which is flat
           > fiterror        ; create error column
           > notice, *       ; for all data:
           > model, lin+gauss; apply complex model
           > fit             ;   for a fit.
                     -> fit exhibits a goodness of fit for the
                        complex model
Revision History
           $Id: cafe_fiterror.pro,v 1.12 2004/05/27 07:33:38 goehler Exp $

Last modified by pro2html on 2005 January 04 at 16:04 UTC

[Home Page] [Software, Documentation] [IDL Documentation] [Quick Reference] [Feedback]

Jörn Wilms (wilms@astro.uni-tuebingen.de)
Updated automatically