PROC IML;
RESET NONAME;
START BP(A) GLOBAL (NY,NCOL,RMEAN,VARX,N);
RMEAN=J(NCOL(NY),1,0);
Q1=RMEAN;
Q3=RMEAN;
RMAD=RMEAN;
MADN=RMEAN;
LOWFEN=RMEAN;
UPPFEN=RMEAN;
LEFT=RMEAN;
F=1;
M=0;
DO J=1 TO NCOL(NY);
SAMP=NY[J];
L=M+SAMP;
TEMP=A[F:L];
Q=quartile (TEMP);
Q1[J]=Q[2,1];
Q3[J]=Q[4,1];
RMAD[J]=MAD(TEMP,"MAD");
MADN[J]=MAD(TEMP,"NMAD");
LOWFEN[J]=Q1[J]-(1.44*MADN[J]);
UPPFEN[J]=Q3[J]+(1.44*MADN[J]);
LEFT=TEMP[LOC(LOWFEN[J]<TEMP & TEMP<UPPFEN[J])];
M=L;
F=F+SAMP;
END;
PRINT LEFT;
FINISH;
********cubaan menggunakan data yg dijana*******;
NY = {11 11 11};
A = {5,8,7,3,9,4,3,29,5,6,7,
3,2,6,4,14,4,7,6,9,3,4,
9,9,8,7,10,11,27,12,15,17,16};
RUN BP(A);
QUIT;
Here, I wish to print matrix free from outliers for three group as the LOWFEN and UPPFEN are the fences to detect outliers and coding LEFT suppose the matrix that have been free from outliers.