Formatting the scree plot x-axis in proc factor

[CBF]

I want to produce a scree plot that I can easily paste into a word document that has an x-axis with only 15 factors. I have tried multiple methods to do this. When I tried ODS RTF, the files were blank. I finally settled on using ODS HTML, but I have not been able to adjust the x-axis on my plot. I tried using a template, but I don't think it is being applied in PROC FACTOR. I tried using SGRENDER, but the images were blank. When I do produce an image, with the code below, the x-axis shows 60 factors, so it is difficult to see where the elbow is. Any help would be greatly appreciated!

PROC TEMPLATE; 

DEFINE STATGRAPH SCREEPLOT; 

NOTES "SCREE PLOT FOR EXTRACTED EIGENVALUES";

BEGINGRAPH / DESIGNWIDTH=DEFAULTDESIGNHEIGHT; 

ENTRYTITLE "EIGENVALUE ((*ESC*){UNICODE LAMBDA}) PLOT"; 

LAYOUT OVERLAY / YAXISOPTS=(LABEL="(*ESC*){UNICODE LAMBDA}") 

XAXISOPTS=(LABEL="FACTOR NUMBER" 

LINEAROPTS=(TICKVALUELIST=(1 2 3 4 5 6 7 8 9 10))); 

SERIESPLOT Y=EIGENVALUE X=NUMBER / DISPLAY=ALL; 

ENDLAYOUT; ENDGRAPH; END; RUN;

ODS GRAPHICS ON ; 

%LET OUTDIR=[PATH] ;

ODS HTML PATH="&OUTDIR" (URL=NONE)

BODY='BODY2.HTML' 

/*CONTENTS='CONTENTS2.HTML'

FRAME='FRAME2.HTML' */ ;

PROC FACTOR DATA = DATA METHOD=PRINCIPAL PRIORS=SMC RESIDUAL  

ROTATE=VARIMAX OUTSTAT=OUTSTAT 

PLOTS(UNPACK)=SCREE REORDER  ;

  VAR &VARS ; 

RUN; 

ODS HTML CLOSE ;

[ballardw]

Did you look at the Proc Factor option NPLOTS= to limit the number of factors plotted?

[Ksharp · Accepted solution]

You could save these data firstly ,afterwards polt it by PROC SGPLOT.

ods select none;
ods output ScreePlot=ScreePlot;
PROC FACTOR DATA = sashelp.heart METHOD=PRINCIPAL PRIORS=SMC RESIDUAL  
ROTATE=VARIMAX OUTSTAT=OUTSTAT 
PLOTS(UNPACK)=SCREE REORDER  ;
  VAR _numeric_ ; 
RUN; 
ods select all;


%let n=6;  /*the number of factors you want to display*/
proc sgplot data=ScreePlot(where=(number<=&n.));
series x=number y=Eigenvalue/markers;
run;

Ksharp_0-1712284711376.png

Or you could check the next graph "the explained variance"  to find elbow point ?

Ksharp_1-1712284861249.png

[CBF]

This worked! Thank you!