Get Date Column

[omega1983]

  I wrote a proc sql to pull in ID, NAME, LS_DATE1 through 4

proc sql;

create table Test as

select ID,NAME,LS_DATE1,LS_DATE2.........etc

from my_table;

quit;

Some rows will have the LS_DATES populated (Not All).  Each row will usually only have one date in one of the LS_DATE columns.  If there are multiple dates in multiple columns in the same row, I want the last date that appears. I want to get the date and place it in its own GET_DATE column with the same value.  I will use the GET_DATE in calulating the difference in days later.

So I need two examples, one that assumes only one date per column per row and another that looks for the last date in a row if there are multiples.

SAMPLE

ID                                      NAME               LS_DATE1            LS_DATE2             LS_DATE3             LS_DATE4

1                                      Smith                 4/15/12

2                                      Bryant                                              1/12/13

3                                      Jones                  5/1/13

4                                     Thomas                                                                            12/13/06

[Haikuo]

If those dates are numeric, use coalesce(), otherwise use coalescec():

proc sql;

create table Test as

select ID,NAME,coalesce(LS_DATE4,LS_DATE3, LS_DATE2,LS_DATE1) as GET_DATE /*put the last date first*/

from my_table;

quit;

HTH,

Haikuo

[omega1983]

data Test

set Test1;

Get_Date=coalesce(LS_DATE1

LS_DATE2 LS_DATE3 Ltr_Dt 😞

run

How can I do the same thing except in a dataset.  All these variables are dates

[Haikuo]

If you need a fast pass for solving your imminent issues, you already have one. If you want to learn how it works inside-out, now it is good for you to spend sometime reading the manual:

SAS(R) 9.3 Functions and CALL Routines: Reference

[NishunkSaxena]

This is a new function for me Coalasce for numeric and coalescec for string. Read articles realted to it on google and to be frank this is a very helpful function for me......Thanks guys