st: -lmoments- revised on SSC

["Nick Cox" <n.j.cox@durham.ac.uk>]

Thanks to Kit Baum as ever, the -lmoments- package on SSC has been
revised. 

L-moments are very widely known in some parts of statistical science,
and, it seems, hardly known in others. The executive summary in the
package description is 

lmoments generates L-moments and derived statistics. L-moments offer 
a systematic, satisfying and indeed elegant approach to the 
summarization of theoretical distributions and data samples and have 
applications in the estimation of parameters and of quantiles, and 
in hypothesis tests for probability distributions.
See Hosking, J.R.M. 1992. Moments or L moments? An example
comparing two  measures of distributional shape. American
Statistician 46, 186-189.  This is version 5.0.0 of the software.
Stata 10 is required for lmoments and for an ancillary command lmose. 
(Previous versions are also included as lmoments9, lmoments8 and 
lmoments6, for which the minimum requirements are Stata 9, 8 and 
 6 respectively.)

The principal changes in this version are as follows. 

1. More of the calculation has been re-written in Mata. The resulting
speed-up is usually 3 or 4 times faster, sometimes more than 10 times
faster. 

2. A standard error calculation has been devolved to an ancillary
program -lmose-. 

3. Both these programs require Stata 10. 

4. Older versions of -lmoments- for version 9, 8 and 6 are included in
the package for anyone who wants to make use of them. 

There are also various minor improvements, at least in my view, in both
functionality and documentation. 

Apologies to anyone inconvenienced by some small syntax changes, but I
do believe this is a better program. 

Nick 
n.j.cox@durham.ac.uk 


*
*   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/