Hello, I'm a student in the accounting department and I'm trying to do some calculations using SAS for the first time.  Actually, I would like to ask for some help with Regression calculation.  I have calculated the real earnings management and now I'm looking for a formula to calculate the REG.  I will be very happy to have all of you as tutors to learn more and get used to the program. Thank you very much.   libname user 'C:\CRISIS' ;            * b1=total assests,  b2=receivables, b3=provision, b4=inventory, b5=tangible assets, b6=land, b7=impairment, b8=subsidiary,               b9=asset in construction, b10=impairment, b11=subsidiary, b12=liability                i1=sales  i2=cost of goods sold  i3=sales and administrative expense  i4=tax and utility  i5=depreciation               i6=rent   i7= insurance  i8=net income, i9=EPS, c1=operating cash flow ; data XXX ;   infile 'C:\CRISIS\KSP97.csv' dlm=',' firstobs=3  lrecl=20000 ;   input code  yr mon  icode opn $ big b1-b12  i1-i9  c1  ;        if b1=0  or i1=0  then delete ;        if mon=12 ; run; * The data of the same year of a firm is deleted ; proc sort nodupkey data=XXX  out=XXX dupout=OO ;  by code yr ;  run; * duplicated data go into OO ; * INDUSTRY NUMBER  ; proc sort data=XXX ; by yr  icode ; run; proc means data=XXX  n  noprint ;         * here,  noprint is very important   ;                 var  mon  ;                 class  yr  icode   ;                 output out=yyy  n=indnum   ;  * n indicates the industry number  ; run; data  indnum ;  set yyy ;                 if yr=. or  icode=.  then delete ;                 keep yr  icode   indnum ; run; *  MERGE financial raw data with industry number ; proc sort data=XXX ; by  yr icode   ; run; proc sort data=indnum ; by yr  icode ; run; data XX ;  merge  XXX  indnum;  by yr  icode;              if indnum < 15  then delete; run; /* Variables Calculation for Real Earings Models  */ proc sort data=XX ; by code yr ;  run; data VAR1 ;  set XX ;             * b1=total assests,  b2=receivables, b3=provision, b4=inventory, b5=tangible assets, b6=land, b7=impairment, b8=subsidiary,               b9=asset in construction, b10=impairment, b11=subsidiary, b12=liability                i1=sales  i2=cost of goods sold  i3=sales and administrative expense  i4=tax and utility  i5=depreciation               i6=rent   i7= insurance  i8=net income, i9=EPS, c1=operating cash flow ;         pcode1=lag(code) ;  pcode2=lag2(code) ; pb1=lag(b1) ; pb2=lag(b2);  pb4=lag(b4);  pi1=lag(i1) ;  ppi1=lag2(i1);   ds1=i1-pi1 ;  ds2=pi1-ppi1 ;   db2=b2-pb2 ;   db4=b4-pb4; pc=i2+db4 ;  de=i3-i4-i5-i6-i7 ; * Variables for real earnings models  ; y1=c1/pb1 ;  x1=1/pb1 ; x2=i1/pb1;  x3=ds1/pb1 ; x4=ds2/pb1 ;  x5=pi1/pb1 ; y2=pc/pb1 ;  y3=de/pb1 ; if code=pcode1=pcode2 ; run; *proc print ;  run ; /* Estimation of REM  */ *outliers are winsorized by year and industry   ; proc means data=VAR1  p1 p99  noprint ;          var  y1-y3 x1-x5 ; CLASS  yr icode ; output  out=outlier     p1=  p1_y1    p1_y2    p1_y3    p1_x1   p1_x2   p1_x3    p1_x4   p1_x5                                      p99= p99_y1  p99_y2  p99_y3  p99_x1 p99_x2  p99_x3  p99_x4  p99_x5 ; run; data  OUTLIER ;  set  OUTLIER ;          if yr=.  or  icode=.  then delete ;         drop _type_  _freq_ ; run; proc sort data=OUTLIER ; by yr icode ; run; proc sort data=VAR1 ; by yr icode ; run; data VVV  ;       *  1% of upper and lower outliers are winsorised  ;     merge VAR1  OUTLIER ;  by yr icode ; if  y1 < p1_y1  then y1=p1_y1 ;  if  y1 > p99_y1 then y1=p99_y1 ; if  y2 < p1_y2  then y2=p1_y2 ;  if  y2 > p99_y2 then y2=p99_y2 ; if  y3 < p1_y3  then y3=p1_y3 ;  if  y3 > p99_y3 then y3=p99_y3 ; if  x1 < p1_x1  then x1=p1_x1 ;  if  x1 > p99_x1 then x1=p99_x1 ; if  x2 < p1_x2  then x2=p1_x2 ;  if  x2 > p99_x2 then x2=p99_x2 ; if  x3 < p1_x3  then x3=p1_x3 ;  if  x3 > p99_x3 then x3=p99_x3 ; if  x4 < p1_x4  then x4=p1_x4 ;  if  x4 > p99_x4 then x4=p99_x4 ; if  x5 < p1_x5  then x5=p1_x5 ;  if  x5 > p99_x5 then x5=p99_x5 ;         run; /*  Estimating Real Earnings Model by year and Industry  */ proc sort data=VVV ;   by yr  icode ; run; proc reg data=VVV  noprint outest=coef1;            /*  Abnormal CFO  */   model  y1= x1-x3  ;   by yr  icode;             run; data coef1; set coef1;   rename intercept=al1  x1=be11 x2=be21  x3=be31  ;   keep yr  icode intercept  x1-x3  ;   run; *proc print; run; proc reg data=VVV  noprint outest=coef2;          /*  Abnormal  production cost  */    model  y2= x1-x4 ;   by yr  icode;               run; data coef2; set coef2;   rename intercept=al2  x1=be12  x2=be22  x3=be32  x4=be42  ;   keep yr  icode intercept  x1-x4 ;   run; *proc print; run; proc reg data=VVV  noprint outest=coef3 ;         /*  Abnormal  discretionary expense  */   model  y3= x1 x5  ;   by yr  icode;              run; data coef3; set coef3;   rename intercept=al3  x1=be13  x5=be53 ;   keep yr  icode intercept  x1 x5 ;   run; *proc print; run; /*  Calculating real earnings management  */ proc sort data=coef1;  by yr  icode; run; proc sort data=coef2;  by yr  icode; run; proc sort data=coef3;  by yr  icode; run; proc sort data=VAR1 ;  by  yr  icode ;  run ; data REM ;       * Here, REM are calculated using VAR01, not VVV, a dataset for estimation   ;         merge  VAR1  coef1  coef2  coef3 ; by yr icode;           acfo=y1-(al1+be11*x1+be21*x2+be31*x3) ;    apc=y2-(al2+be12*x1+be22*x2+be32*x3+be42*x4) ;    ade=y3-(al3+be13*x1+be53*x5) ;  rem=acfo*(-1)+apc+ade*(-1) ; run; proc sort ; by code yr ; run; proc print; run; quit ; ... View more