eprintid: 21473
rev_number: 12
eprint_status: archive
userid: 2326
dir: disk0/00/02/14/73
datestamp: 2016-07-08 11:32:18
lastmod: 2016-07-21 10:16:24
status_changed: 2016-07-08 11:32:18
type: workingPaper
metadata_visibility: show
creators_name: Sawitzki, Günther
title: Testing Numerical Reliability of Data Analysis Systems
subjects: 510
divisions: 110400
keywords: Regression, Statistical computing, Software quality control
cterms_swd: Regressionsanalyse
note: Erschienen in: Computational Statistics and Data Analysis 18.2(1994), S. 269-301
abstract: From 1990 to 1993, a series of tests on numerical reliability ofdata analysis systems has been carried out. The tests are based onL.Wilkinson's "Statistics Quiz". Systems under test included BMDP, Data Desk,Excel, GLIM, ISP, SAS, SPSS, S-PLUS,STATGRAPHICS. The results showconsiderable problems even in basic features of well-known systems. For allour test exercises, the computational solutions are well known. The omissionsand failures observed here give some suspicions of what happens in lesswell-understood problem areas of computational statistics. We cannot takeresults of data analysis systems at face value, but have to submit them to alarge amount of informed inspection. Quality awareness still needs improvement.
date: 1994-07
id_scheme: DOI
id_number: 10.11588/heidok.00021473
schriftenreihe_cluster_id: sr-10b
schriftenreihe_order: 01
ppn_swb: 1657867307
own_urn: urn:nbn:de:bsz:16-heidok-214736
language: eng
bibsort: SAWITZKIGUTESTINGNUM199407
full_text_status: public
place_of_pub: Heidelberg
pages: 36
citation:   Sawitzki, Günther  (1994) Testing Numerical Reliability of Data Analysis Systems.  [Working paper]     
document_url: https://archiv.ub.uni-heidelberg.de/volltextserver/21473/1/report.01.pdf