Dear All,   I am trying to save parameters as macro value. I have a variable called beta with values 1-8, represented beta1 - beta8. I would like to save each betaas as a macro value called 1bar-8bar, but I am having trouble with the naming of the variable 'ibar' in the symput statement:    How could I achieve this or similar result?   *store the estimated parameters; data _null_; set t0; do i= 1 to 8; if beta = i then call symput('ibar', Estimate); end; run; ... View more

Dear All,   I am trying to save parameters as a macro: data t0; length Parameter $15; input Parameter $ Estimate; datalines; Intercept 1132 ; run; *store the estimated parameters; data _null_; set t0; if Parameter = "Intercept" then call symput('Ibar', Esimate); run; %put &Ibar;     I get the message that IBar was not saved properly: SYMBOLGEN: Macro variable IBAR resolves to .   Why can"t I save the value?     ... View more

Dear sas-users,   How would I go about setting the confidence-intervals to 99% in the following code?   proc sgplot data=claims; title 'Simple Linear Regression Fit Plot -- PROC SGPLOT'; reg y=indavgcost x=year / cli clm; xaxis values=(2010 to 2013 by 1); run;     Thanks ... View more

      Dear SAS-users,   I am running several regressions as above. I am creating 'parmest' dataset to save the results. My problem is that the length of the variable "parameter" in parmest_1 and parmest_2 are different due to the max. length of the parameter. One has 15 characters the other one has 19. This means that the name on the categorical variable "year" will be different also: "year 2010" vs 'year    2010'.   When I merge the two datasets ( parmest_1 and parmest_2), SAS doesn't recognize them as the same name.   Somehow I need to set the length of the variable parameter even before it is created. How would I go about doing that?   Thanks.           data claims_1; input client year cost contract; datalines; 18 2011 589.92 10.2 19 2010 1629.8 11.5 20 2011 1813.29 2.3 21 2012 412.06 3.2 22 2012 219.82 0.4 23 2010 3669.98 5.9 24 2012 4879.63 6.8 ; run; /*model 1*/ proc glm data=claims_1; class year (ref='2010' ); model cost=year/ solution; ods output ParameterEstimates=work.parmest_1; run; /* model 2*/ proc glm data=claims_1; class year (ref='2010' ); model cost=year|contract/ solution; ods output ParameterEstimates=work.parmest_2; run; data work.parmest_1; set work.parmest_1; keep Parameter Estimate; rename Estimate=Estimate1; run; data work.parmest_2; set work.parmest_2; keep Parameter Estimate; rename Estimate=Estimate2; run; proc sort data=parmest_1 out=parmest_1; by Parameter; run; proc sort data=parmest_2 out=parmest_2; by Parameter; run; DATA work.parmest; length parameter $19; MERGE parmest_1 parmest_2; BY Parameter; RUN; ... View more

Dear SAS users,     I am trying to use PPOC REPORT with ORDER. The input dataset (parmest) is sorted already by numord. I want the output table (TABLE_SHORT) to be reported in order of numord, but it is coming out in completely different order.  See input and output datasets attached.   What am I doing wrong?     proc report data=parmest nowd out=table_short; column variable GROUPS ('Total-Model 1' Estimate1 StdErr1 Probt1) ('Total-Model 2' Estimate2 StdErr2 Probt2) ('Total-Model 3' Estimate3 StdErr3 Probt3) ('Gen. Hosp.-Model 3' Estimate4 StdErr4 Probt4) ('UMC Hosp.-Model 3' Estimate5 StdErr5 Probt5) ('TC Hosp.-Model 2' Estimate6 StdErr6 Probt6) ('TC Hosp.-Model 3' Estimate7 StdErr7 Probt7) ; define groups/group noprint; define Variable /order order=data display; define estimate1 - estimate7 /order order=data display 'Estimate'; define StdErr1 - StdErr7/order order=data display format=stderrf.; define probt1 - probt7 /order order=data display; run;     ... View more