Hi. I have the following 5 datasets. Each has a variable named review_ind, which can be either 1 or 0.  Can I merge the datasets together, but only keep the rows if review_ind=1 in at least one of the datasets.   Is there an easy/efficient way to do this during the merge or will I need to individually subset each of the 5 datasets (where review_ind=1) before I merge?  I'm trying to reduce the overall lines of code in this program.   Any suggestions would be greatly appreciated.   DATA ALL_CLAIMS; MERGE &USERN._LIB.F3_DISTANCE_CAR_F &USERN._LIB.F4_RFRG_NVR_BLG_CAR_F &USERN._LIB.F5_BENE_NO_HIST_CAR_F &USERN._LIB.F6_BENE_SHARE_CAR_F; BY ID_NUM BENE_SK; RUN; ... View more

Hi.  I have two different datasets named my_data and other_data.  Both datasets have an id variable that is 10 char.  The first dataset's id variable is named my_id and the later is other_id.    I'm wondering what is the best way to query a list of ids that appear in both datasets?   Any help would be greatly appreciated.  Thanks. ... View more

Hi. I have the following cards statement, but I can't figure out how to read in the date column (CL_FROM_DT) as a SAS date instead of a character string.  I'd appreciate the help.  Thanks.   DATA HAVE; LENGTH ID_NUM $ 10 ENRLMT_NUM $ 9 CL_FROM_DT $ 9; INPUT ID_NUM $ B_ID $ ENRLMT_NUM $ CL_FROM_DT $ H_CD $; DATALINES; 1174620710 1169339 030490008 30JAN2018 L0180 1174620710 1169339 030490008 30JAN2018 L3960 1285980003 1169339 454072764 30DEC2017 L0650 1477921666 1169339 472633814 23JAN2018 L1851 1477921666 1169339 472633814 23JAN2018 L2397 1851421663 1169339 561727024 22DEC2016 L0627 1851421663 1169339 561727024 01MAY2018 L0976 1851421663 1169339 561727024 01MAY2018 L1030 1851421663 1169339 561727024 01MAY2018 L1040 1851421663 1169339 561727024 01MAY2018 L1050 1851421663 1169339 561727024 01MAY2018 L2397 1851421663 1169339 561727024 01MAY2018 L2810 1366995144 2318543 813131377 08DEC2017 L0650 1366995144 2318543 813131377 08DEC2017 L3916 1417255761 2318543 274931342 20NOV2017 L0650 1972863868 2318543 455151476 14NOV2017 L0650 ;RUN;   ... View more

Unlike a similar question you guys just helped me with - about how to de-dup a list of comma delimited values stored in a variable.  My challenge is time is to find a method to prevent the CATX function from ever putting a duplicate date value into the list to begin with.  If you have any thoughts on how to alter the code below to accomplish this I'd really appreciate the insights.     The following piece of SAS code takes data rows that looks like this:   BENE   FROM_DT (stored in a SAS dataset as SAS date values) 12345  23OCT2010 12345  11NOV2015 12345  11NOV2015 12345  15DEC2011 12345  11NOV2015   And changes the multiple rows into a single row by Bene with the dates all transformed into a single variable named From_Date_List using the CATX function:   BENE   FROM_DT_LIST 12345  23OCT2010, 11NOV2015, 11NOV2015, 15DEC2011, 11NOV2015        PROC SORT DATA=FROM_DT_RESHAPE; BY BENE_SK; RUN;   DATA FROM_DT_RESHAPED; LENGTH FROM_DT_LIST $1000; DO UNTIL (LAST.BENE); SET FROM_DT_RESHAPED; BY BENE; FROM_DT_LIST = CATX(', ',FROM_DT_LIST, PUT(FROM_DT, DATE9.)); END; KEEP FROM_DT_LIST BENE; RUN; ... View more

Hi.  I'm wondering if there's a way to de-dup a list of values stored in a variable.  There could be duplicates in the middle or end of the string and there could be multiple duplicates within the string.  Any ideas for doing this? I'd really appreciate the help.  Thanks.   HAVE_ID_NUM 1033112065, 1629420120, 1033112065, 1629420120, 1033112065, 1033112011, 1629420120   WANT_ID_NUM 1033112065, 1629420120, 1033112011 ... View more

Hi.  I'm wondering if there isn't a simpler way to accomplish what my code below is doing. For each record on my table there is a CLM_BLG_PRVDR_NPI_NUM and a CLM_RFRG_PRVDR_NPI_NUM.  My requirement is to assign a Flag Indicator = 1 to each record where the associated CLM_RFRG_PRVDR_NPI_NUM value is not also the same value as ANY record's  CLM_BLG_PRVDR_NPI_NUM value on the entire table. Otherwise the Flag Indicator should be 0.   As you can see from my code below I accomplish this in two Proc SQL steps.  I'm interested to know if there is a simpler or more efficient way to get the same results in just one Proc SQL step?   Any insights would be much appreciated!     /* CREATE LIST OF RFRG NPI WHO HAVE NEVER BEEN A BLG NPI AND ASSIGN 1 AS A FLAG INDICATOR */ PROC SQL; CREATE TABLE RFRG_NEVER_BLG AS SELECT DISTINCT CLM_RFRG_PRVDR_NPI_NUM ,1 AS FLAG_IND FROM NFPP_TRD.&USERN._ORTHOTICS_BASE_PULL_2 EXCEPT SELECT DISTINCT CLM_BLG_PRVDR_NPI_NUM ,1 AS FLAG_IND FROM NFPP_TRD.&USERN._ORTHOTICS_BASE_PULL_2 ;QUIT;   /* MERGE ON THE RFRG NPI WHO HAVE NEVER BEEN A BLG NPI AND SET AN INDICATOR FLAG */ PROC SQL; CREATE TABLE F4_CL_ATRISK_V1 AS SELECT T2.* ,CASE WHEN T1.FLAG_IND=1 THEN 1 ELSE 0 END AS FLAG_IND FROM BASE_PULL_2 AS T2 LEFT JOIN RFRG_NEVER_BLG AS T1 ON T1.CLM_RFRG_PRVDR_NPI_NUM = T2.CLM_RFRG_PRVDR_NPI_NUM ;QUIT; ... View more