Last edited on 20th April, 2004 by gs/ gs. - Home page: WWW StatLab Heidelberg or CD home.

Diagnostic Plots for One-dimensional Data

This is a loose leaf collection of work sheets on common (or not so common) diagnostic plots, built up to help in every day consulting. The collection is presented in as an active document, allowing to use your own data or your simulations.
It is included as one chapter in the graduate course "Data Analysis and Regressiondiagnostics" and accessible in full on the course CD.

random samples

If you have Java 1.2 installed, this is an interactive image. Click the upper half to get a random new random number generator - and a random new sample size. Click the lower half to see the current distribution. This applet will not work with Java 1.0; sorry.

Video: video Real or Windows media.

Background discussion (PDF): Keeping Statistics Alive in Documents

For those interested in statistics:

This collection has been originally published as: G. Sawitzki: Diagnostic Plots for One-Dimensional Data. in: P.Dirschedl & R.Ostermann (eds.) Computational Statistics. Papers collected on the occasion of the 25th Conference on Statistical Computing at Schloss Reisensburg. Heidelberg, Physica, 1994, ISBN 3-7908-0813-X, pp. 234-258.
See the PDF version of the paper. If you want to use any of the methods, download one of the shrink-wrapped versions of the paper and experiment on your local machine, using simulations or your own real data sets.

Downloading:

A new release is available upon request.

For those interested in computing:

The paper has been implemented using the Voyager technology, as documented in: G. Sawitzki:Extensible Statistical Software: On a Voyage to Oberon. Journal of Computational and Graphical Statistics Vol. 5 No 3 (1996)
See the PDF version of the paper. To experiment with the software, download one of the shrink-wrapped versions of the paper. The shrink-wrapped versions use a reduced form or Oberon Microsystem's BlackBox framework. If you have a full version of BlackBox installed, move the Onedim subsystem to your version of BlacBox and discard the rest of BlackOne.
Onedim is written in Oberon. The source code is availabe for Oberon/F.
Onedim is not Voyager. It is using a reduced, shrink-wrapped excerpt of Voyager. We use it as an experiment to try various forms of packing and distribution. So we are very interested on any feedback whether you can easily download an use it, or whether any problems occured. For the main stream of Voyager, see the Voyager home page.

A video on the software technology is available: video Real or Windows media.
For download, use http://www.statlab.uni-heidelberg.de/projects/onedim/onedim.mov or http://www.statlab.uni-heidelberg.de/projects/onedim/onedim.mp4.

For a background discussion on some didactical aspects of integrated documents, see G. Sawitzki: Keeping Statistics Alive in Documents Preprint: Sonderforschungsbereich 373 Quantifikation und Simulation Ökonomischer Prozesse, Berlin 1999, to appear in Computational Statistics.

various diagnostic plots

Known problems

Mathematical formulae do not have a portable format. Formulae may be corrupted if some fonts are missing.