For i:12 you can do
-1 | 2 1 0
-0 | 9 8 7 6 5 4 3 2 1
+0 | 0 1 2 3 4 5 6 7 8 9
1 | 0 1 2
but variants of the stem and leaf display are available, for example
-1 | 2 1 0
* | 9 8 7 6 5
-0 | 4 3 2 1
+0 | 0 1 2 3 4
* | 5 6 7 8 9
1 | 0 1 2
in which final digits 0 through 4 go next to the first digit and final digits 5
through 9 go next to an appropriate * . In the t f s e variant used below,
final digits 0 and 1 go next to the first digit, final digits 2 and 3 go next
to t, final digits 4 and 5 go next to f, final digits 6 and 7 go next to s, and
final digits 8 and 9 go next to e . You try to allocate data 0's evenly to +0
and -0.
If you can get a copy of Understanding Robust and Exploratory Data Analysis
edited by David C. Hoaglin, Frederick Mosteller, and John W. Tukey, Chapter 1
Stem-and-Leaf Displays explains all.
--Kip
Sent from my iPad
On Mar 14, 2013, at 2:53 PM, Raul Miller <[email protected]> wrote:
> On Thu, Mar 14, 2013 at 3:47 PM, km <[email protected]> wrote:
>> Classical stem and leaf uses truncation not rounding, to make it easy to
>> look up values in the original data. For the wiki data
> ...
>> You cannot show the symmetry in i: 12, which classically would probably
>> appear as (again done by hand, t f s e are mnemonics for two-three four-five
>> six-seven eight-nine)
>>
>> t | 2
>> -1 | 1 0
>> e | 9 8
>> s | 7 6
>> f | 5 4
>> t | 3 2
>> -0 | 1
>> +0 | 0 1
>> t | 2 3
>> f | 4 5
>> s | 6 7
>> e | 8 9
>> 1 | 0 1
>> t | 2
>
> I do not understand how we have any stems other than _1 0 1 for i:12.
> I can see a case for -1 and -0 for small negative values, but I do not
> understand what you've displayed here.
>
> --
> Raul
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
