MathGroup Archive 1996

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: {mg3340] quartile

  • To: mathgroup at smc.vnet.net
  • Subject: [mg3373] Re: {mg3340] quartile
  • From: Lou Talman <me at talmanl.mscd.edu>
  • Date: Sat, 2 Mar 1996 04:14:20 -0500
  • Sender: owner-wri-mathgroup at wolfram.com

Bryce Kelly wrote:

> On Mathematica's definition of a quartile.
> 

> Hello All,
>
> I need some help on the definition of a quartile. I have run into the
> following problem on different definitions for quartiles between Maple
> and Mathematica .
>
> Using Mathematica 2.2.1
> In[] := <<Statistic`DescriptiveStatistics`
> In[] := data = {10,20,30,40,50,60,70,80}
> In[] := Median[data]
> Out[] = 45
> In[] := Quartiles[data]
> Out[] = {25,45,65}
> 

> Using Maple V release 4
> > data := [10,20,30,40,50,60,70,80]:
> > with(stats):
> > with(describe):
> > median(data);
> 		45
> > quartiles := [seq(describe[quartile[i]],i=1..3)]:
> > quartiles(data);
> 		[20,40,60]
> 

> What is Mathematica source for the definition of a quartile ?
> 

One might better ask for the source of Maple's definition of "quartile".  I've  
never seen a definition of "quartile" that made the second quartile different  
from the median.  And most folks I know would agree that the first and third  
quartiles should be the medians of the lower and upper halves of the data--as  
Mathematica appears to think.

--Lou Talman

==== [MESSAGE SEPARATOR] ====


  • Prev by Date: composites
  • Next by Date: Re: {mg3340] quartile
  • Previous by thread: Re: composites
  • Next by thread: Re: {mg3340] quartile