Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: New commands -dmout- and -pctrim-


From   Michael Barker <[email protected]>
To   [email protected]
Subject   st: New commands -dmout- and -pctrim-
Date   Mon, 3 Feb 2014 10:16:08 -0500

Thanks to Kit Baum, two new commands are available on the SSC archive:
-dmout- and -pctrim-

DMOUT: module to create difference-in-means tables.
dmout produces difference-in-means tables. Means of each variable in
varlist are compared across the values of the by() variable. dmout is
useful when comparing several variables across two groups, such as
treatment and control groups in an RCT, or attriters and non-attriters
in longitudinal data.

PCTRIM: module to trim variables based on percentiles.
pctrim trims outlying observations based on percentile bounds. pctrim
can operate on a varlist. The user can create an indicator variable
marking outliers or recode them. Recode options include mean, median,
upper/lower bounds, or system missing.

pctrim is similar to the recent SSC addition, -winsor2-, by Lian
Yu-jun. Both commands allow users to trim or winsorize several
variables at once. The main difference is in their treatment of
missing values. -winsor2- operates on each variable independently. The
default behavior for pctrim first identifies the analysis sample, with
no missing data, then operates on that sample only. pctrim can operate
on variables independently with the missok option. Both pctrim and
winsor2 are related to the original -winsor- command, by Nicholas J.
Cox.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index