BookmarkSubscribeRSS Feed
Gayatriv
Calcite | Level 5
I am using following syntax to create compare ROC curves using SAS:
Code for multiple and Single ROC :
----------------------------------------------------------------------------------------------------------------------
FILENAME REFFILE 
'/folders/myfolders/sasuser.v94/ExcelData/ACPC/finalMCI_AD_ninsstat.xlsx';

PROC IMPORT DATAFILE=REFFILE DBMS=XLSX OUT=WORK.IMPORT;
GETNAMES=YES;
RUN;

PROC CONTENTS DATA=WORK.IMPORT;
RUN;

ODS GRAPHICS ON;

PROC LOGISTIC DATA=WORK.IMPORT PLOTS(ONLY)=ROC(ID=OBS);
MODEL GROUP(EVENT='0')=ACC_GSH_Conc PCC_GSH_Conc CINGULATE_GSH_Conc;
ROC 'ACC' ACC_GSH_Conc;
ROC 'PCC' PCC_GSH_Conc;
ROC 'CINGULATE' CINGULATE_GSH_Conc;
RUN;

PROC LOGISTIC DATA=WORK.IMPORT PLOTS(ONLY)=ROC(ID=OBS);
MODEL GROUP(EVENT='0')=ACC_GSH_Conc;
RUN;

ODS GRAPHICS OFF;
 
Why SAS is giving two different values for single ROC curve? When I am plotting single ROC for 'alb' AUC value is different than I got by the syntax given above for single ROC (roc 'Albumin' alb;) with multiple ROC.
1 REPLY 1
StatDave
SAS Super FREQ

As was suggested in your tracking entry on this question, you should use the NOFIT option in the MODEL statement to remove the second AUC estimate that is from the MODEL statement.

Ready to join fellow brilliant minds for the SAS Hackathon?

Build your skills. Make connections. Enjoy creative freedom. Maybe change the world. Registration is now open through August 30th. Visit the SAS Hackathon homepage.

Register today!
What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 1 reply
  • 382 views
  • 2 likes
  • 2 in conversation