to find out more. I will send a copy of the new version to SSC when Kit
Baum is back in Boston after 26 August.
The -parmest- package is described as below on my website. The new version
should do exactly the same as the old version, but a bit faster, as I have
streamlined the code to eliminate unnecessary -preserve- statements. The
new version took 38% less time than the old version when tested on large
datasets containing 100 to 2000 copies of the -auto- data concatenated end
to end as by-groups, with one additional by-variable added. However, even
in the new version, execution time is still quadratic in the number of
by-groups, because -parmby- still loads each by-group into the memory using
-restore,preserve- and dropping all by-groups except one. Any further
improvement would probably require a total rewriting of the code, similar
to what is being done with -statsby-, only a bit more radical. I am
considering how this might be done (probably with the Stata 8 -postfile-),
and am keeping watch on further developments with -statsby-.
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
package parmest from http://www.kcl-phs.org.uk/rogernewson/stata8
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
TITLE
parmest: Convert estimation results to a data set with 1 obs per
parameter
DESCRIPTION/AUTHOR(S)
parmest creates an output data set, with 1 observation per parameter
of the
most recent estimation results, and variables corresponding to
parameter names,
estimates, standard errors, z- or t-test statistics, P-values,
confidence
limits and other parameter attributes. parmby is a quasi-byable
extension to
parmest, which calls an estimation command, and creates a new data
set, with 1
observation per parameter if the by() option is unspecified, or 1
observation
per parameter per by-group if the by() option is specified. The
output data
set may be listed to the Stata log and/or saved to a file and/or
retained in
memory (overwriting any pre-existing data set).
Author: Roger Newson
Distribution-Date: 18august2003
Stata-Version: 8
INSTALLATION FILES (click here to install)
parmest.ado
parmest.hlp
parmby.ado
parmby.hlp
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
--
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]
Website: http://www.kcl-phs.org.uk/rogernewson
Opinions expressed are those of the author, not the institution.
