Hello, I'm new at this, so not sure if I'm explaining this correctly, but I'm trying to estimate predicted probably of a doctor having enough interaction with a patient based on predictors having a set value.  So probabilty Doctor 1, Doctor 2, Doctor 3, etc  has enough interaction with the patient if the patient is age 18-34, female, with a college degree or higher.  I tried using this code below using "estimate", but it doesn't seem to be working.  Anyone know what I might be doing wrong?   proc glimmix data=tables2 method=laplace noclprint; class attending; model doc_interaction (descending) = age18_34 age35_54 gender_m edu_hs_m edu_some_m edu_col_M /CL DIST=Binary link=logit solution; estimate "Intercept Age 18-34, female, college grad or higher" int 1 age18_34 1 gender_m 1 edu_col_m 1 / ILINK; random intercept / subject=attending S CL Type=VC ; COVTEST / wald; run;   ... View more

I created a state map using the following code below, but now I'd like to export the graph as a high quality image in png or jpeg format.     proc gmap data=graph map=graph all resolution=6 ; id state county_codex; choro region / discrete legend=legend2 coutline=black anno=annoreg; legend1 shape=bar(4,3) pct across=5 frame cshadow=black label=none; legend2 label=(position=top j=center h=1 position=(top) color=black "Region Names") position=(bottom left outside) shape=bar (4,3) value=(c=black j=l h=1) frame across=5 cborder=black ; run;   The sas output in SAS looks like this However, when I add the following code before running proc gmap in an attempt to export the map as a png, the image of my map suddenly shrinks and my legend enlarges.     FILENAME plots "C:\Users\"; goptions reset=all DEVICE=png300 transparency GSFNAME=plots GSFMODE=REPLACE NOFILEONLY; ods _all_ close; ods listing; is there some other code I can use to simply export the file as I see it in my sas output screen?   ... View more

Is there an easier way to specify a range when your variable is a character?   For example, I am recoding a bunch of revenue/procedure codes into categories, but using the IN() function doesn't work well for ranges that are not numeric.  So I would like to do this:   if c_dtl_proc_cd in (59000:59899, 90801:90899, A4651:A4932) then category='2. Other'; But this doesn't seem to work.  It seems like I would have to do this:   if (c_dtl_proc_cd ge '59000' and c_dtl_proc_cd le '59899') or (c_dtl_proc_cd ge '90935' and c_dtl_proc_cd le '90899') or (c_dtl_proc_cd ge 'A4651' and c_dtl_proc_cd le 'A4932') then category='2. Other';  Which wouldn't be too bad, except that some of the categories have a series of different ranges. Is there an easier way to list these ranges?   Thanks! ... View more

I have a macro that reads through a list of states and years, however, I'm wondering if there's a more efficient way to read through the states instead of only using a macro.   Here is my code:   %macro test (st=,year=); proc freq data=data.out_&st._&year.; tables place; run; %mend test; %test (st=NY, year=2011); %test (st=NY, year=2012); %test (st=NY, year=2013); %test (st=VA, year=2011); etc. Is there an easier, more consise way to run through the years without having to have three lines for NY for each year?  Should I use an array? ... View more