Welcome to SAS forum. Typical example of a case where you would use Proc format aka user defined format and use a put function or format statement in a datastep to apply that format.  ... View more

data have; infile datalines truncover; input period $30. ; datalines; Sunday, June 4, 2017 Monday, June 5, 2017 Tuesday, June 6, 2017 Wednesday, June 7, 2017 Thursday, June 8, 2017 Friday, June 9, 2017 Saturday, June 10, 2017 Sunday, June 11, 2017 Monday, June 12, 2017 Tuesday, June 13, 2017 Wednesday, June 14, 2017 Thursday, June 15, 2017 Friday, June 16, 2017 Saturday, June 17, 2017 Sunday, June 18, 2017 Monday, June 19, 2017 Tuesday, June 20, 2017 Wednesday, June 21, 2017 Thursday, June 22, 2017 Friday, June 23, 2017 Saturday, June 24, 2017 Sunday, June 25, 2017 Monday, June 26, 2017 Tuesday, June 27, 2017 Wednesday, June 28, 2017 Thursday, June 29, 2017 Friday, June 30, 2017 Saturday, July 1, 2017 Sunday, July 2, 2017 Monday, July 3, 2017 Tuesday, July 4, 2017 Wednesday, July 5, 2017 Thursday, July 6, 2017 Friday, July 7, 2017 Saturday, July 8, 2017 Sunday, July 9, 2017 Monday, July 10, 2017 Tuesday, July 11, 2017 Wednesday, July 12, 2017 Thursday, July 13, 2017 Friday, July 14, 2017 Saturday, July 15, 2017 ;   data want; set have; array wkday(*)  Sunday      Monday     Tuesday    Wednesday       Thursday  Friday      Saturday (7*0); do _n_=1 to dim(wkday); if vname(wkday(_n_))=scan(period,1) then wkday(_n_)=1; else wkday(_n_)=0; end; run; ... View more

@Theo_Gh OK, I will try to work on your stuff. However, Today I have got lectures and some studying to do for my exam next week. If time permits, I will seriously help you out, but please do not totally rely on me. I am sure some Super user will respond with a code solution regardless.  ... View more

You are welcome. ; is basically known as a null statement. It's very interesting statement and has some relevance as it is an executable statement. In some instances, you can even use that for conditional processing however that is out of scope for this thread. Have fun! ... View more

@subrat1  If you attach file, honestly I don't think it is convenient for any of us to help. Super user @RW9 has often many a time wrote requesting people to post in the form of a datastep and I wonder not many people are doing that.   Anyway, the below code works for your sample:     data want; if (_n_ = 1) then do;    if 0 then set have;     declare hash myhash(dataset: "have(where=(Scenario='BUDGET'))");     myhash.definekey('Region','Country','Product','SubProduct','Quarter' );     myhash.definedata('Gross_Premium_Written','Net_Premium_Written');     myhash.definedone();  end;  set have(where=(Scenario='ACTUALS') rename=(Gross_Premium_Written=_Gross_Premium_Written Net_Premium_Written=_Net_Premium_Written));  rc=myhash.find(); if rc=0 then do; Scenario='DELTA'; Gross_Premium_Written=_Gross_Premium_Written-Gross_Premium_Written; Net_Premium_Written=_Net_Premium_Written-Net_Premium_Written; output; end; drop rc _:; run;         ... View more

Oh right, just the placement of the ; should have been after the last dataline in the next line ... View more

Data colors ; input Color : $1. @@ ; datalines ; R R B G Y Y . . B G R B G Y P O O V V B ;/*notice here*/   what's the problem ... View more

Please give us a sample of your Input1- Yours data with 40 countries and your look up dataset that has exchange rate daily. The immediate approach comes to my mind is load the exchange rate table in hash table and one pass on your data looking up the hash table should be a breeze ... View more

data have ; input id date1 : date9. res date2 : date9.; format date1 date2 date9.; cards; 111 10AUG2017 10 10AUG2017 111 10SEP2017 20 10SEP2017 111 11SEP2017 30 11SEP2017 111 09MAR2018 18 09MAR2018 ; run;     data want; do _n_=1 by 1 until(last.id); set have; by id; k=lag1(res); k2=lag2(res); if _n_=1 then do; output; continue; end; else if _n_=2 then mean=k; else mean=sum(k,k2)/2; output; end; drop k:; run;   Regards, Naveen Srinivasan ... View more