Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: -zmap- command available on SSC


From   "Nick Cox" <[email protected]>
To   <[email protected]>
Subject   st: -zmap- command available on SSC
Date   Tue, 16 Mar 2010 13:20:59 -0000

Thanks to the ever efficient and energetic Kit Baum, a -zmap- command is
now available from SSC. Stata 8 is required. 

Use -ssc- to install or inspect if interested. 

-zmap- is a fairly obvious convenience command that was developed for
teaching use with spatial series and tested on a remotely sensed image
dataset with almost 600,000 observations with distinct x and y
coordinates and various responses (z) measured for each. At that scale
the areas (pixels, or more precisely their ground equivalents) can be
considered as point locations for graphics and indeed the only possible
marker symbol is a point.  

Be that as it may, some people may be interested in its uses for other
kinds of trivariate data. The help file carries a detailed worked
example in which other marker symbols are advisable.

A bit more formally: -zmap- graphs (or maps) binned values of a variable
z with respect to two variables x and y treated as Cartesian
coordinates. The range of z is
divided into two or more bins or classes and points in each bin are
shown distinctly. The resulting plot is thus a composite scatter plot.
The user may specify percentile breaks or breaks on the scale of z.

I'm a bit surprised to be thinking that no-one has published precisely
this kind of program before, but apologies if I'm overlooking something.
Several people must have done this kind of thing ad hoc, however. 

Maurizio Pisati's much more versatile -spmap- I understand to be based
on the idea of mapping areas as polygons, which is not needed here. 

Sergiy Radyakin gave a talk at the DC conference last summer with what
at first sight some similar (and very impressive) results. But as the
underlying program and help files appear to remain private I cannot
comment in detail, except that Sergiy uses what he calls "advanced Stata
programming", which I don't do.  

Vince Wiggins made a very helpful suggestion on getting (colours of)
markers to show visibly in the legend. In essence, each variable shown
by a point has a bigger sibling with the same colour but a square
symbol, which features in the legend but not in the plot (and vice
versa). 

Nick 
[email protected] 


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index