>> > >> >> The following is extracted from an article by Jon Honeyball in PC
>>Pro,
>> > >> >> Issue 62, December 1999, pp 248-255.
>> > >> >>
>> > >> >> "In front of me right now is a paper entitled On the accuracy of
>> > >> >> statistical procedures in Microsoft Excel 97, reprinted from the
>>Journal
>> > >> >> of Computational Statistics and Data Analysis, which is a highly
>> > >> >> prestigious, refereed academic journal. I might without
>>exaggeration
>> > >> >> call it 'the bible of computational statistics' and there's arguably
>> > >> >no higher reference in the world. The article comes from volume 31,
>> > >> >issue 1, 28 July 1999.
>> > >> >>
>> > >> >> The abstract for the paper says: 'The reliability of statistical
>> > >> >> procedures in Excel are assessed in three areas: estimation (both
>> > >> >> linear and nonlinear), random number generation, and statistical
>> > >> >> distributions (such as for calculating p-values). Excel's
>>performance in all three
>> > >> >> areas is found to be inadequate. Persons desiring to conduct
>> > >> >> statistical analyses of data are advised not to use Excel.' As an
>> > >> >> opening statement, you must admit that it's a bit of a corker.
>> > >> >>
>> > >> >> The paper's authors, BD McCullough and Berry Wilson of the Federal
>> > >> >> Communications Commission in Washington DC, go on to describe in
>> > >> >> precise detail how they applied the recently released StRD
>>(Statistical
>> > >> >> Reference Datasets) from the American National Institute of
>>Standards
>> > >> >> and Technology to assess the performance of Excel in a wide range of
>> > >> >> statistical tests. The results are stunningly bad, and, worse
>>still, the paper refers
>> > >> >> back to work done by Sawitski in 1994 on Excel 4 and the problems
>> > >> >> reported then are still present in Excel 97. I've run some of the
>>tests myself and
>> > >> >> they're still there in Excel 2000. The paper, which can't really be
>> > >> >> argued with, is littered with phrases like 'can be judged
>>inadequate'
>> > >> >> and 'it can be deduced that Excel uses an unstable algorithm'. The
>> > >> >> authors find fault with its univariate summary statistics, analysis
>>of
>> > >> >> variance, linear regression, nonlinear regression, random number
>>generation and
>> > >> >> so forth. What can I say? If you use the statistics add-on package
>>that
>> > >> >> ships with Excel, you really better know your stuff because Excel
>>may
>> > >> >> well come up with wrong numbers.
>> > >> >>
>> > >> >> Excel's statistics add-on pack is riddled with potential disaster
>> > >> >> areas, and since it has been subjected to the best analysis
>>available
>> > >> >> i n the world and found to be wholly lacking, the only applicable
>>words
>> > >> >> are 'avoid' and 'plague'. Instead, you should buy yourself a
>>decent stats
>> > >> >> add-on package that has numerical methods that are open to peer
>>review
>> > >> >> and whose authors know what they're doing (unfortunately,
>>Microsoft's
>> > >> >> stats-pack team obviously doesn't).
____________________________________________________________________
Mark David Major
Lecturer, University College London
Secretariat, International Space Syntax Steering Committee
Contact Address at Work:
The Bartlett School of Graduate Studies
(Torrington Place Site)
University College London
Gower Street
London WC1E 6BT
United Kingdom
tel +44 (0)20 7679 5916
fax +44 (0)20 7916 1887
email [log in to unmask]
Contact Address at Home:
8 Crediton Hill
West Hampstead
London NW6 1HP
United Kingdom
tel +44 (0)20 7431 5278
fax +44 (0)20 7431 8576
email [log in to unmask]
[log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|