LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (June 2003, week 3)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:   Wed, 18 Jun 2003 09:55:14 -0400
Reply-To:   "Fehd, Ronald J. (PHPPO)" <rjf2@CDC.GOV>
Sender:   "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:   "Fehd, Ronald J. (PHPPO)" <rjf2@CDC.GOV>
Subject:   Re: Outliers
Comments:   cc: Aj <ajaya_u2002@YAHOO.COM>
Content-Type:   text/plain

> From: Aj [mailto:ajaya_u2002@YAHOO.COM] > I am trying to find the ways to find outliers in data. I have > used proc univariate option. > > I would like to know simple options where I can find easily. > I would appriciate your time.

SAS Books By User author Ron Cody has a book which explains how to use various procedures to produce your -- what's the right word here?: threshhold or limit --

basically you run a proc assign the limit to a macro variable and then check your data for rows above/below the limit

keyword: algorithm method OOP return an object

something like:

%LET LIBREF = LIBRARY; %LET DATA = MYDATA; %LET VAR = Var-A; %LET OP = lt;%*less than; %LET OP = le;%*less than or equal; %LET OP = gt;%*greater than; %LET OP = ge;%*greater than or equal;

proc SOMETHING data = &LIBREF..&DATA.; *number-crunching happens here to &VAR.; out = LIMIT (keep = Limit);

%*assign Limit to macro variable; DATA _NULL_; set LIMIT; call symput('LIMIT',Limit); stop;run; %*or; proc SQL noprint; select Limit into :LIMIT from WORK.LIMIT ;quit;

%PUT LIMIT<&LIMIT.>;

PROC PRINT data = &LIBREF..&DATA. (where = (&VAR. &OP. &LIMIT. )); TITLE "&LIBREF..&DATA. &VAR. &OP. &LIMIT."; run;

for a summary report of values in a data set that are not defined by formats see my paper

NESUG 2001 Adv Tutor pg 53 %INVALID: a data review macro using proc FORMAT option other=INVALID to identify and list outliers http://www.pace.edu/nesug/proceedings/nesug01/at/At1008.pdf

for use of statistical procedures see my colleague Handsfield's paper:

check our most excellent archives: http://www.listserv.uga.edu/archives/sas-l.html search for: subject contains: outliers author's address: RJF2 since: Jan 2003 until: Jan 2003

Ron Fehd the macro maven CDC Atlanta GA USA RJF2@cdc.gov


Back to: Top of message | Previous page | Main SAS-L page