Hi all, I have successfully created an annotated state map from geocoded address info using the states map. However, when I try to use the county map data set, it appears that the geocoding output does not match  what information in the county map dataset. Here is a small program to illustrate the difficulty. Code doesn't get past proc gproject. It seems that the length of x and y of the map do not match the length of x and y in the geocoded output. thanks! Phil : libname lookup "D:\Documents\desi2023\frlanalyses\geocodedata__2006__ZIP4_Geocode_Data\data"; data zips; input c zip; anno_flag=1l cards; 1 65203 2 65211 3 65201 ; proc geocode /* Invoke geocoding procedure */ method=plus4 /* Specify geocoding method */ lookup=lookup.zip4 /* Lookup data from MapsOnline */ data=zips /* Input data set to geocode */ out=geocodedzips; /* Specify name of Output data set of locations */ run; data mocounties;set maps.counties;if state=29; data mocounties;set mocounties geocodedzips;run; proc gproject data=mocounties out=MOCountiesP; id state ; run; data my_map locations; set mocountiesP; if anno_flag=1 then output locations; else output my_map; run; data anno_locations; set locations; if c=1 then color="red"; if c=2 then color="bib"; if c=3 then color="viyg"; if c=4 then color="bippk"; xsys='2'; ysys='2'; hsys='3'; when='a'; function='pie'; rotate=360; size=.5; style='psolid'; output; length html $300; html='title='||quote(trim(left(citystate))||'0d'x||trim(left(put(zip,z5.)))); style='pempty'; color='deb'; output; run; pattern1 v=e; proc gmap data=mymap map=mymap anno=anno_locations; id county; choro segment / coutline=black levels=1 nolegend coutline=gray99;run; ... View more

I recently started using the TableN macro and I wanted to say thank you to @JeffMeyers for providing such a useful macro! I was able to produce Table 1 for TM_group (which is a binary variable with values 0,1). However, I would now like to create Table 1 for subgroups based on the ADRD_group variable (which is also 0,1). The documentation for the macro indicates the COLBY option can do exactly what I'm looking for, and provides an example output table with Gender and Arm. However, I can't find good example syntax of how to add the COLBY option to the code. I've tried a few variations, but I keep getting errors. Below is my latest version.   %tablen( data=have; ,by=TM_group .colby=ADRD_group ,var=DEM_AGE DEM_SEX ,type=2 2 ,outdoc=C:\Users\r\Desktop\TM vs MA ADRD Table 1(&version.).rtf ); ... View more

Hello,   I am getting this error message for INPUT function, could anyone please help me with this issue? TIA. Here is my code: data lb; merge adsl(in=a) lb1(in=b); length AVALC $50. SUBJID $15.; by USUBJID; if a and b; INDEX+1; SUBJID=substr(USUBJID,13,7); AVAL=LBSTRESN; AVALC=LBSTRESC; if LBCAT='COAGULATION' and LBTESTCD in ('APTT' 'INR' 'PT') or LBCAT='URINALYSIS' and LBTESTCD in ('URBILV' 'URGLUV' 'URPHV' 'URPROTV' 'URSGV'); run; proc sort data=lb1(where=(index(TRT01A,'Placebo')>0 and LBDTC^='')) out=lb_1(where=(input(LBDTC,e8601dt16.)<=TRTSDTM)); by INDEX; run;   I'm getting ERROR: INPUT function reported 'ERROR: Invalid date value' while processing WHERE clause. LBDTC is character 2023-11-07T12:50  ... View more

When I subset a dataset in a DATA step, SAS will continue to run for much longer than expected, so long in fact that I have not seen it finish running. However, when I break the run and cancel the submitted statements, the log indicates that 1,271 observations were read, which is number of observations that I expect to have in the subset. Why is it that SAS keeps running when all of the observations that match the WHERE condition have been read?   In the DATA step I use a WHERE statement to subset for observations where the character variable SUB = '123'. The dataset is large (1.3M+ obs.), but as I mentioned, the resulting data set "filtered_items" should only have 1,271 observations.  libname corpxin "\\filepath\folder"; data filtered_items; set corpxin.items_202001; where SUB = '123'; run; ... View more

I have a exported an excel report that contains empty cells and would like to write a code that would color all empty cells in the excel output. I have the following code which only highlights/colors empty cells in the NAME column, however I'd like to highlight all empty cells without having to write out code for all 47 columns that I have in my report :   PROC REPORT DATA = STAGE1; DEFINE NAME -- REFERRAL_BASIS/ DISPLAY; COMPUTE NAME; IF NAME= '' THEN CALL DEFINE('NAME', "STYLE", "STYLE=[BACKGROUND=CYAN]"); ENDCOMP; RUN; ... View more