I would like to have an SAS code for PCA analysis. I would appreciate if you could also send an example input file and output result file.
Hi,
You can find good description with good examples in the following linke:
http://jonathantemplin.com/files/multivariate/mv05psyc990/psyc990_10.pdf
http://www.ats.ucla.edu/stat/sas/output/principal_components.htm
http://support.sas.com/documentation/cdl/en/statug/63347/HTML/default/viewer.htm#statug_factor_sect028.htm
I suggest SAS analyst for PCA. It produces nice graphs.
http://support.sas.com/documentation/cdl/en/anlystug/58352/HTML/default/viewer.htm#chap13_sect2.htm
Resulted code for example:
*----------------------------------------------------------+
| Generated: Wednesday, August 7, 2019 10:18:31 |
| Data: C:\Users\Admin\Documents\My SAS Files(32)\9.2\new2 |
+----------------------------------------------------------*;
title ;
footnote;
*** Principal Components Analysis *** ;
options pageno=1;
proc princomp data=SASUSER.new2;
var ALKPH NEFA BHBA GUL CHOL HDL TRG LDL VLDL TPRO CA P MG GGT ALB OX BCS
KETOSIS;
run; quit;
goptions reset=all device=WIN;
*** Principal components plot ***;
proc princomp data=SASUSER.new2 out=work._scores std noprint;
var ALKPH NEFA BHBA GUL CHOL HDL TRG LDL VLDL TPRO CA P MG GGT ALB OX BCS
KETOSIS;
run;
title;
footnote;
proc corr noprint data=work._scores out=work._corr;
var ALKPH NEFA BHBA GUL CHOL HDL TRG LDL VLDL TPRO CA P MG GGT ALB OX BCS
KETOSIS prin:;
with ALKPH NEFA BHBA GUL CHOL HDL TRG LDL VLDL TPRO CA P MG GGT ALB OX BCS
KETOSIS;
run;
data work._scores;
set work._scores(in=inscore)
work._corr(where=(_type_='CORR'));
if inscore then do;
_type_ = 'SCORE';
_name_ = 'ROW'||left(put(_n_,12.));
end;
run;
/* --- put # of PRIN# variables in macro variable --- */
data _null_;
dsid = open("work._scores");
_maxn_ = 0;
do _i_ = 1 to 9999;
if varnum(dsid,'PRIN'|| left(trim(put(_i_,4.)))) > 0 then
_maxn_ = _i_;
else leave;
end;
call symput('_maxprin',left(trim(put(_maxn_,4.))));
rc = close(dsid);
run;
options nodate nonumber;
/* --- macro definition --- */
%macro _myplot(plotvar1,plotvar2,desc);
%let dim1 = %substr(&plotvar1,5);
%let dim2 = %substr(&plotvar2,5);
%if &dim1
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Modeling
- Analyses of Variance