Re: Candidate stem-and-leaf plot function

[Daniel J Sebald]

On 01/26/2013 12:23 AM, Michael D. Godfrey wrote:
> On 01/26/2013 01:12 AM, Daniel J Sebald wrote:
> > I had to change "signbit" to "sign" and use 1/-1 in place of 0/1,
> > respectively. Is "signbit" your own routine? Or am I missing
> > something? That change would be necessary if placed in the base scripts.
> Just quickly, signbit is in the current development system 3.7.1+. I
> assume that this
> would be released before 3.8.

I thought I had something fairly recent, but maybe something was added 
to the repository recently.  Thanks.


> Also, the sort order for negative stems is, I think, the way it should be,
> i.e. the same order as the data.

Oh, I didn't look closely to see that neither are the leaves on other 
stems in a sorted order.  The wikipedia page suggests the data should 
always be sorted as part of the plot.  Maybe "sorted" and "unsorted" 
could be options with the default being whatever is most common.


> By the way, I just checked the R function. It looks reasonable, but they
> chose
> to call it just stem(). The is a lot shorter than stem_and_leaf_plot, so
> I think changing
> the name would make sense. I cannot remember why I chose such a long name.

"stem" is already used in Matlab/Octave.  How about "stemandleaf"?  Or 
maybe "stemleaf"?  "sal"/"salplat" gets too arcane I think.


> And, I will think about how to show the plot in a plot window. The
> original idea
> was to just use the symbols available on typewriters. Not so relevant
> today.

We'll see if others on the list have ideas for plotting ASCII characters 
graphically.  There may be a slick way to do so in gnuplot, but that 
isn't generally applicable I suppose.

You mention in the comments of the m-file about scaling could be more 
complex.  It seems to me such a plot could be generalized in terms of 
number base in order to get good granularity in both directions.  The 
left (stem) column could be base-16.  It's just that may not naturally 
fit into the "standard" range of a stem and leaf plot.  The user can 
scale the data and remove any bias, but then that sort of makes things a 
little more difficult to interpret.

Dan