[R] [OT] Oo-calc & StAtistics

[(Ted Harding)]

On 17-Jun-10 20:36:27, Marc Schwartz wrote:
> On Jun 17, 2010, at 3:03 PM, Ted Harding wrote:
>> The thread "R licensing query" currently running has raised
>> the classic critcisms of using Excel for statistics.
>> 
>> I was wondering: Has anyone applied the same or similar set
>> of tests to OpenOffice "calc"?
>> 
>> Or would the Executive Summary be: "Calc is just like Excel"?
>> 
>> (Not that I'm a spreadsheet user, if I can avoid it; but I
>> sometimes get asked about such things).
>> 
>> Sorry to be completely non-R here, but I can't think of a
>> better place to ask!
>> 
>> Ted.
> 
> Ted,
> The most recent work on this that I know of, that includes Calc, is:
> 
> On the Numerical Accuracy of Spreadsheets
> Marcelo G. Almiron, Bruno Lopes, Alyson L. C. Oliveira, Antonio C.
> Medeiros, Alejandro C. Frery
> JSS
> Vol. 34, Issue 4, Apr 2010
> http://www.jstatsoft.org/v34/i04
> 
> There have been some discussions on R-Help in the past (circa 2003),
> which I have participated in, which discuss some of the faults in Calc,
> such as rounding numbers "close to 0" to 0 (like Excel does) and some
> of the non-IEEE 754 floating point behavior (like Excel does)...
> 
> HTH,
> Marc Schwartz

Thanks, Marc, it helps indeed! Not so off-topic as I feared, either,
since R itself was used as a tool in the investigations, and gets
honourable mention at the end:

  "Finally, as a rule of the thumb, every user should be aware
   that spreadsheets have serious limitations. Other platforms
   are advisable, being currently R the most dependable FLOSS
  (Free/Libre Open Source Software, see Almiron et al. 2009)."

Executive Summary: There is no general justification at all for
using oocalc rather than Excel. It is sometimes better, sometimes
worse, does not have the same features as Excel, and in its own
way can drop you just as deeply in it.

Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding at manchester.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 17-Jun-10                                       Time: 22:29:44
------------------------------ XFMail ------------------------------