Isnt the code dsd dlm = ' | '    the method to help avoid excess characters, in this case excess | when using proc export to .txt. Here is my fields in the dataset LOAN       CLOSING_DATE     DEFAULT_REASON  LAST_NAME    EMAIL Desired output 111111119|05/15/2014|Layoff|Shelton|shelton@gmail.com Instead I get this 111111119|05/15/2014||||||||||||Layoff|||||||||||||| Shelton|shelton@gmail.com I get extra  | and the word jump to the next line.  The Loan number should always begin the proc export. ... View more

data step_1300 step_1302; set my_steps; if step_cd='1300' then output m_1300; else if step_cd='1303 then output m_1302; run; data m_1300; set m_1300; 1300_dt=compl_dt; step1300=step_cd; format 1300_dt mmddyy10.; run; data m_1302; set m_1302; 1302_dt=compl_dt; step1302=step_cd; format 1300_dt mmddyy10.; run; data merge; merge m_1300 m_1302; by loan_num; date1=max(1300_dt.1302_dt); run; sample output Loan_Num       date1                    step_cd 00000              6/15/2014               "the max step cd according to the step date" Here is my question. I can get the max compl_dt as shown here.  I need to be able to also show the max step (step1300 or step1302  and apply this to multiple dates and step cd) I am just using two here as example.  So if step1300 ended up being the max date I want to display in a char column '1300' ... View more

data mydata(keep=name address city state zip phone email); set data1; run; I am using a proc export proc export data=mydata; set mydata1; File "/servername/report..txt" Linesize=8000 dsd dlm='|'; run; In some cases there may be a blank in one or more fields Osborne 1313 E Main St Washington DC myemail@me.com  (In this case we skipped the zip and phone number In my proc export  I am getting double || and sometimes I get what appear to be small boxes.  The carraige returns are not working.  The display should begin a new line after email however in some cases a new record will begin on the same line Osborne 1313 E Main St Washington DC myemail@me.com Bradeley 1312 E Main Would a trim function fix this in some way? ... View more

Here is the code data test; set test1; if date ge anydate; run; proc format $cfmt  '075'='fs075'                            '079'='fs079'                           '080'='fs080' ; run; proc transpose data= test out = test2  format step_cd $cfmt. id step_cd; by ln_no; var anydate; run; proc sql; create table test4 as select a.ln_no, b.fs075, b.fs079, b.fs080 from mytable  a left join test2  b on a.ln_no=b.ln_no; quit; This works when the fs varaibles have data.  When they are blank or null, I get errors.  The vairable is not found or in correlation with b.  How can I account for the variables in my code if they happen not to have values on a particular day?  I this case the highlighted variables of fs075 079 or 080?? ... View more

DATA _NULL_;     CALL SYMPUT('date1',&prev_me);   RUN;   CALL SYMPUT('date1',&prev_me); SYMBOLGEN:  Macro variable PREV_ME resolves to 201405 The &prev_me is part of an internal calendar that automactically formats the year previous month, in this case May. Now I need to format this date in this format  month year  (may14)  or the text version of the month then the year. This will allow me to use this to capture datasets based on their month and year (ie. link_may14_he How can I adjust the data null to achieve this result. ... View more

Data test(keep=dept  hire_dt  company  department); Set test2; Run; Data _Null_;   Set test2;     File "/xxs11/xx/sascode/xx/out/xx_&pbus_1._&Out_NAME..txt"         LINESIZE=8000 dsd dlm='|';           PUT &Final_VARS ;              run; produces a txt file like this 11111|05/15/14|xyz|enginerring problem is I am getting carraige return issues 11111|05/15/14|xyz|enginerring|22222|12/11/2014|xyz|accounting The output is supposed to stop and jump to the next line adter department How can I adjust the code in data_null to address this.  I have heard of @@  and missover ... View more

data want; set have; if step=1 then step1 =completion_dt; if step=2 then step2=completion_dt; if step=3 then step3=completion_dt; run; proc sql; create table want1 as select a.ln_no, a.address, a.city, b.step1, b.step2, b.step3 from main  a inner join  want b on a.ln=b.ln; quit; Here is the issue.  If say step3 has no value then the proc sql statement fails because it is a null value.  How can I address this when the value is null however I still need to join the two datasets ... View more

Here is my scenario Table1 Loan1            Loan2     Loan_Pos    Name 1111              1112            1             Geter 1112              1111            2             Geter   move this one to table 2 1113               1115           1             Snow 1116                                 2             Baldwin Table 2 Loan1            Loan2     Loan_Pos    Name 1112              1111            2             Geter   move this one to table 2    Here is the issue.  Loan_Pos pertains to the loan position for loan1.  For example 1111 is in a 1st position and 1112 (same row) is the 2nd.  The second row has the same information except in reverse.  1112 is the second position and 1111 is the first.  Therefore the row = 2 becaise the second loan is in the left most position.  In that scenario I need for the loan to be placed in another table.  Now for 1116 it only has a second loan so it stays in the first table What makes this tricky is you are analyzing 2 consecutive rows that are reversing loan numbers.  I hope this visual is sufficient to advise on how to accomplish this movement of the loan       ... View more

Here is a sample output loan_number                                             Past_Due_This_Month            Past_Due_Mar2014    Past_Due_Feb2014      Past_Due_Jan2014   Past_Due_Dec2013  and back 1 yr.......... 1111                                                                     2                                         2                  1122                                                                                                                2                                                                      2 1145                                                                   2                                                                                                                   2                          2 I need to keep records that have the months past due fields populated with a  minimum of 2  for a minimum of 2 consecutive months.  So in this case loans 1111 and 1145 would qualify. 1122 would be deleted.  How would I do this ... View more