Thanks to Kit Baum, the latest version of -smileplot-, with Stata 8
graphics, is now downloadable from SSC. In Stata, type -ssc desc smileplot-
to find out more.
The -smileplot- package takes, as input, a data set with 1 observation per
estimated parameter and data on P-values and the corresponding parameter
estimates, as might be created by -parmest-, -parmby-, -statsby- or
-postfile-. It carries out one of a selection of multiple test procedures
on the P-values, and expresses these multiple procedures graphically by
drawing smile plots. A smile plot plots the P-values (on a reverse log
scale) on the Y-axis against the estimates (or another variable) on the
X-axis, with reference lines on the Y-axis corresponding to the uncorrected
and corrected critical P-values. In the new version, the program
-smileplot- has been upgraded to produce the new Stata 8 graphics. However,
the program -multproc- is still Stata 7, and a new program -smileplot7- has
been added for users who still want to draw smile plots using the old
quick-and-dirty Stata 7 graphics. (For what it's worth, Stata 7 users can
also still download the old Stata 7 version of -smileplot- from my website at http://www.kcl-phs.org.uk/rogernewson/stata7/
and this might possibly reduce the amount of program revision required if
and when they upgrade to Stata 8.)
I plan to present some examples of graphs from the new -smileplot- package
at the 9th UK Stata User Meeting in London on 20 May. Thanks in advance to
Kit Baum (again), my Adobe Acrobat presentation should soon be available
for download on the RePEc page for the meeting at http://ideas.repec.org/s/boc/usug03.html
along with downloadable material from the other presenters.
The new -smileplot- package does not yet have a dialog interface. This is
going to be a more difficult project, because I am working under Windows
98, which (as b. water and Jean Marie Linhart were saying only yesterday)
is not an easy system to run advanced dialog boxes under. So,
unfortunately, users who want their smile plots to look exactly right will
still have to know to specify a -size()- suboption for title sizes, a
-labsize()- suboption for axis tick label sizes, and a -mlabsize()-
suboption for data point label sizes. However, this is of course not the
fault of anybody at StataCorp. Thanks again to all at StataCorp for giving
us all a beautiful graphics system.
Best wishes
Roger
--
Roger Newson
Lecturer in Medical Statistics
Department of Public Health Sciences
King's College London
5th Floor, Capital House
42 Weston Street
London SE1 3QD
United Kingdom
Tel: 020 7848 6648 International +44 20 7848 6648
Fax: 020 7848 6620 International +44 20 7848 6620
or 020 7848 6605 International +44 20 7848 6605
Email: [email protected]
Opinions expressed are those of the author, not the institution.
