Hello everyone,   I have a database with values saved as character. The number of decimal points can vary between observation I.e. both 10,132 and 6,254252 could be found. I want to convert these character values to a numeric value while retaining all decimal points. Using input(char,best.) yields me 1 decimal points. Simply multiplying the character string with 1 yields me 5. Exactly what is going on here, and how can i adjust the code to include all/more decimals? See example below.   data test; format CharVar $50. value_num1 value_num2 20.14; CharVar="7663229328,8184132475796"; Value_num1 = input(tranwrd(CharVar,",","."), best.); Value_num2 = tranwrd(CharVar,",",".")*1; run; CharVar= 7663229328,8184132475796 Value_num1=7663229328.800000000 Value_num2=7663229328.818410000 ... View more

hello, I have a  question about running connect to option  in PROC SQL. Currently im able to do calculations as such: proc sql feedback; connect to odbc as DB (dsn="[name]"); create table work.Want as select * from connection to DB ( select t1.*, divide(t2.A,t2.B) as ABC from DB.[Scheme].[Have_on_server] AS t1 left join work.[have_in_work] AS t2 on t1.A=t2.A ); disconnect from DB ; quit; Where i take a dataset from server, left join a dataset from my work and put the dataset in my work. But if i instead take a dataset from work and left join a dataset on server, this does not work. Does anyone have a clue as of why it isnt working? proc sql feedback; connect to odbc as DB (dsn="[name]"); create table work.Want as select * from connection to DB ( select t1.*, divide(t2.A,t2.B) as ABC from work.[Have_in_work] as t1 left join DB.[Scheme].[Have_on_server] AS t2 on (t1.A=t2.A) ); disconnect from DB ; quit; ... View more