If the Data Quality Server is licensed at your site then you could use function DQSTANDARDIZE() SAS(R) 9.3 Data Quality Server Reference Else: It's quite a bit of coding and if you search this forum then you'll find multiple discussions around data cleansing and standardisation.
... View more
Hi, ANYDTDTEw. Reads and extracts the date value from any of the following: DATE, DATETIME, DDMMYY, JULIAN, MDYAMPM, MMDDYY, MMxYY*, MONYY, TIME, YMDDTTM, YYMMDD, YYQ, YYxMM*, month-day-year Try anydtdte11. in your case.
... View more
Here is some code that looks to work OK if you want to do it yourself. data tov7; input name $char32.; v7name = name; nl = length(v7name); if notfirst(first(v7name)) then do; if length(v7name) lt 32 then v7name = '_'||v7name; else substr(v7name,1,1)='_'; end; do until(f eq 0); f = findc(v7name,,'NKT'); if f then substr(v7name,f,1)='_'; end; do until(l = k); l = length(v7name); v7name = transtrn(v7name,'__','_'); k = length(v7name); end; cards; hospital type: local Jun-30-2005 100 blank first. lots of blanks 12345678901234567890123456789012 ;;;; run; proc print; run;
... View more
Similar to those above- using data step rather than SQL but still calculating year difference rather than the simple incrementation of Mit data have; input patient $ year doctor $; cards; A 1992 Doctor_1 A 1993 Doctor_1 B 2005 Doctor_3 B 2006 Doctor_2 B 2008 Doctor_2 B 2010 Doctor_2 ;;;; run; proc sort data=have; by patient doctor year; run; data want ; set have; by patient doctor; if first.doctor then tenure=1; else tenure=tenure+(year-lastyear); lastyear=year; retain lastyear tenure; drop lastyear; run; proc sort data=want; by patient year doctor; run;
... View more
I ran the code. It is able to get the file name and the number of observation correct, but all other variables in the individual TSV file is missing. Could you please tell me how would I best modify the code so that it would accommodate .TSV file. The thing is the tab separated file contains "" for each value in the file. Excel can real the file no problem, but maybe not SAS... Thanks -
... View more
