hi, i guess due to below step................the variable 'lagged_doseadj'...will give blank....only.. Check whetherr this is missed in data reading...or in  'have' dataset. data want; set have;   by patient; doseadj=''; lagged_actl=lag(actldose); lagged_doseadj=lag(doseadj); .... ..... ... View more

hi , Try this.. data want (drop =i); informat break_date date9.; format  break_date date9.; do i = 1 to (3*30); Break_date=intnx('days','01Apr2013'd,i); if weekday(break_date) = 1 then break_date=intnx('days',break_date,1); output; i=i+4; end; run; Regards, Allu ... View more

hi, check this....if really want unique 'Codeid' proc sql; select distinct a.codeid ,a.aum, b.date from table_a a inner  join   table_b b on a.codeid=b.codeid and a.date=b.date ; run; Regards Allu ... View more

hi, if you need the differnce  b/n two years you can go with proc compare.... data have; input employement &$28. @29 y_2005 comma7.@37 y_2006 comma7. @45 y_2007 comma7.; format y_2005 comma7. y_2006 comma7. y_2007 comma7.; datalines; in the nursing labour force 253,592 254,956 277,297 employed in Nursing         243,916 244,360 263,331 clinical nurse              203,517 196,998 212,480 ; run; proc compare base=have nosummary out=year_per; var y_2005; with y_2006; var y_2006; with y_2007; run; Regards Allu i think , you can go with with steps ...proc transpose ...and lag () function also ....but please check this .... ... View more

hi , Compress() function will delte the required characters Tranwrd() replaces the words .. please try..... data want;    input description $ 32.;   /* new_description1 = compress(description,'%,&'); for deleing only*/   description=tranwrd(description,'%','Percentage');   description=tranwrd(description,'&','And');    datalines;    TYP/BGS 8X80'S+50% XF    WEL W GRP&PEA 12X1    WENS & CRAN 8X180G    ;    run; Regards Allu ... View more

yes ...i need the Observation values in cloumn wise ............... 1_10   2_10    3_10............ 1         11 2         12   3         13 ...        . .           . .           . 10        20 Regards Allu ... View more

thanks Tom.. ... View more

hi, when you are pasting the data from 'datalines ' in to txt file .....some wrong  alligment for the  'tdate' and 'hourstamp'...is happening .. while importing it will give wrong...allingment...please check this or eg : please delete the first row of your data and check... i have not observed any errors with following code .. Note : i have adjusted the alligment of first row !! data TEMP_DS; infile "C:\Users\galax_allu\Desktop\adjdata.txt" truncover pad; OPTIONS DATESTYLE=MDY;   format timeStamp dateampm22. ;    input tdate mmddyy11. hourStamp adjCount comma8.;   timeStamp=dhms(tdate,hourStamp,30,0);   keep timeStamp hourStamp adjCount; label         timeStamp = "TimeStamp"         adjCount = "Adjucation Count" ; run; Regards Allu .. ... View more

hi , use truncover and pad ...in the follwing step..?? INFILE "/user_data/adjdata.txt"                 truncover                 DSD; input tdate mmddyy11. @13  hourStamp @16 adjCount comma8.; ... View more