Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: New Programs on SSC


From   "Ulrich Kohler" <kohler@wz-berlin.de>
To   statalist@hsphsun2.harvard.edu
Subject   st: New Programs on SSC
Date   Mon, 23 Feb 2004 10:58:19 +0100

hi all,

I have sent the following programs to Kit Baum, asking him to include 
them in the SSC-Archive. I gues that the programs will be available 
in the next few days of so. Comments and bug reports would be highly 
appreciated.

1) biplot.ado
--------------

Biplot.ado is an upgrade of the program originally written for Stata 
5 to Stata 8.2. The upgrade uses the new graphics and adds some  
features to the program. Among the new features are Relative 
Variation Diagrams, Weights, a data transformation which lets you 
interpret the distances in the scatter as mahalanobis distances 
(instead of euclidean) and a possibilty for highlighting observations 
from different subpopulations. 

A more detailed explanation of biplots will be given in my talk on 
the German Stata Users' Meeting on April 5th 2004 in Berlin, and 
shortly afterwards in the proceedings of that meeting in the 
internet. 


2) _gmlabvpos.ado
-----------------

_gmlabvpos.ado is an egen function to automatically generate a 
variable for the clock-positions of marker labels in scatterplots
(something to include as varname in the option mlabvpos(varname))
The general idea of the program is to not allow the marker labels 
interfere with the quantitative information of the marker symbols (as 
proposed by Cleveland 1994, Elements of Graphing Data, pg 43). The 
program does not try, however, to prevent the marker labels from 
overplotting. Therefore the program gives satisfying results only for 
scatterplots with just a few observations.

BTW: I wonder if there exist a more general algorithm for clock-
positions. The problem seems to be similar to the labeling of 
geographical objects in maps, and there might be a solution from  
Geographers which can be applied. However it seems that labels in 
maps are placed by hand. If there is no general algorithm, Stata 
should probably also offer the possibilty for such placements by 
hand. I don't want to have STAGE back, but how about a graphical tool 
to build the necassary -generate- and -replace- commands.  

many regards
uli

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index