Hello, I have a question regarding way to weight population for each individual. I have used syntax below to have fractional rank but found out that I need to weight total population of every insur_level before doing the PROC RANK syntax. I have created the sample data set to let you understand what I mean easier. Could anybody help me 'weighting' the population?       PROC RANK DATA=SSTEMP OUT=TEMP; VAR INSUR_LEVEL; RANKS INSUR_RANK; RUN;   PROC MEANS DATA=TEMP NOPRINT; VAR INSUR_RANK; OUTPUT OUT=INSUR_RANK_MAX MAX=INSUR_RANK_MAX; RUN;   DATA TEMP2; SET TEMP; IF _N_=1 THEN SET INSUR_RANK_MAX; INSUR_RANK_FRAC=INSUR_RANK/INSUR_RANK_MAX; RUN;   ... View more

Hello, I'm writing it to ask for help coding with macro code below. It's my first time seeing macro code and I have to change it to ordinary(non macro) code. Can anybody help me changing this macro code to ordinary sql and data code?   Thank you.   %macro CI1(indata=,xvar=,wealth=,outdata=); /* First step*/ data _null_; set &indata nobs=nobs end=end; if end; call symputx("number",nobs); run; /* Second : ranking wealth*/ proc rank data =&indata out = &xvar.ranka;  var &wealth;  ranks wealthrank;  run; proc means data = &xvar.ranka noprint;  var wealthrank;   /* wealthrank_max*/ output out = wealthrank_max max = wealthrank_max; run;   /* wealthrank _ max，wealthrank_max*/ data &xvar.rank;  set &xvar.ranka ; if _n_ = 1 then set wealthrank_max;  wealthrank_frac = wealthrank/wealthrank_max;  run;   /* Third : covariance*/ proc corr data = &xvar.rank COV outp=cov(type=cov) noprint; var &xvar wealthrank_frac;  run; data cova; set cov; if _name_ ="wealthrank_frac" and _type_="COV"; cov = &xvar;  keep cov;  run;   /* Fourth. Calculating CI*/ proc means data = &indata noprint; var &xvar; output out = mu mean = mu;  run; data mua; set mu; keep mu;  run; data ci; merge cova mua;  CI = 2* COV/mu; run;     data &outdata; length xvar $ 15; xvar="&xvar"; n=&number; set ci(keep=ci); run;   proc datasets lib=work noprint; delete &xvar.: ci cov cova mu mua wealthrank_max ; run; quit; %mend;   ... View more

Hello. I have question regarding concentration index(CI) and Horizontal equity Index(HIwv index) sas code. I have stata code the way to calculate CI(there are 2 ways to calculate CI) and HIwv index but I don't know how to type it in SAS. Could you please help?     ** method 1 (CI) sum outpts sca m_y= r(mean) corr rank outpts [fw=wt], c sca covph2w = r(cov_12) sca ci = 2*(covph2w)/m_y sca list ci ----------------------------------------------------------------------------------------------------- ** method 2 (CI) qui sum rank [fw=wt] sca var_rank = r(Var) ge ihs=2*var_rank*(outpts/m_y) reg ihs rank [pw=wt] sca c=_b[rank] sca list c summarize seq sca sigma = r(Var) gen temp = (2*sigma*inpts) / mean regress temp seq regress temp seq [pw=wt], robust cluster(psu) sca c = _b[seq] sca list c regress temp seq [pw=wt] nlcom ((2*(sigma))/(_b[_con] + 0.5*_b[seq]))*_b[seq] ----------------------------------------------------------------------------------------------------- * HIwv Index   < 2σr2 [yi/μ - hi/η] = α + βRi + μi > xi: zinb outpts i.age_g sex i.edu i.sah, inf(age_g sex i.edu i.sah) robust cluster(psu) predict out_zinb gen util_c = (util_inc > 300) + (util_inc > 700) + (util_inc > 1000) gen sah_3 = (sah > 2) + (sah >3) sort ri gen need = sum(out_zinb) gen cum_need = need / need[_N] label var cum_need "필요의료이용" twoway (line cum_in seq, clpattern(dot)) /// (line seq seq)(line cum_need seq, clcolor(pink)) graph export "2005 외래 이용 HIwv곡선.emf", replace       ... View more