[S] Functions for estimation incorporating sampling weights

Frank E Harrell Jr (fharrell@virginia.edu)
Wed, 29 Apr 1998 13:57:57 -0400


A series of functions for obtaining weighted estimates is now in the
Hmisc library in statlib or our web page. The functions are listed below.
Their names are self-explanatory. wtd.loess.noiter does a 1-iteration
loess smooth efficiently using internals that are used by loess.smooth.
loess.smooth does not allow weights to be passed to the internals.
wtd.loess.noiter is a stripped-down version of loess.raw.

The weights may be integers, representing frequencies used to "magnify"
observations, or fractional numbers such as general sampling weights.
For the latter case, there is an option to make weights sum to the number
of observations, after missing data have been deleted.

Soon I'll incorporate weights in the binary/ordinal logistic regression function
(lrm) in the Design library.

Functions:

wtd.mean
wtd.var
wtd.quantile
wtd.ecdf
wtd.table
wtd.rank
wtd.loess.noiter

---------------------------------------------------------------------------
Frank E Harrell Jr
Professor of Biostatistics and Statistics
Director, Division of Biostatistics and Epidemiology
Dept of Health Evaluation Sciences
University of Virginia School of Medicine
http://www.med.virginia.edu/medicine/clinical/hes/biostat.htm

-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news