create new variable from earliest and latest lagged variables

[aj_goodnews]

Dear all,

I have a data set as follow,

ID	year	lag4_var	lag5_var	lag6_var	lag7_var	treat
1	2000	3	2	6	3	0
1	2001	6	7	8	9	0
1	2002	6	5	7	8	0
1	2003	6	7	11	8	1
1	2004	12	11	10	9	1
2	2015	6	7	8	9	0
2	2016	3	2	6	3	0
2	2017	13	1	5	6	1
2	2018	7	6	8	12	1
2	2019	5	22	5	6	1

in this data set, lag4_var = lag4(var) where var is some measurement of firm size for each ID/quarter. Similarly, lag5_var is the value of the 5^th previous quarterly ‘var’ and lag6_var is the value of the 6^th previous quarterly ‘var’, and so on. Treat is a dummy variable equals 1 if any of lag#_var greater than or equals 10, or equals to 0 if all the lag lag#_var less than 10.

I want to create an new variable, K. For observations with treat=1, K equals the earliest lag#_var that is greater than or equals 10. For observations with treat=0. K equals the latest (most current) lag#_var.

For example, for observation with ID=1 and year=2000, treat=0, and in this case K should be lag4_var as lag4_var is the most recent number for lag#_var.

For observation with ID=1 and year=2004, treat=1, and in this case K should be lag6_var as lag6_var is the first and earliest lag#_var that is greater than 10.

So as a result, I want to have the following data set.

ID	year	lag4_var	lag5_var	lag6_var	lag7_var	treat	K
1	2000	3	2	6	3	0	3
1	2001	6	7	8	9	0	6
1	2002	6	5	7	8	0	6
1	2003	6	7	11	8	1	11
1	2004	12	11	10	9	1	10
2	2015	6	7	8	9	0	6
2	2016	3	2	6	3	0	3
2	2017	13	1	5	6	1	13
2	2018	7	6	8	12	1	12
2	2019	5	22	5	6	1	22

Thanks in advance for your help!

[ErikLund_Jensen]

Hi @aj_goodnews 

Please remember that it is good practice to supply input data as working code to create an input data set. It is a great help for anybody trying to answer your question, because creating test data from text is often the most time consuming part.

The code to solve your problem is simple:

data have;
input ID year lag4_var lag5_var lag6_var lag7_var treat;
cards;
1 2000 3 2 6 3 0
1 2001 6 7 8 9 0
1 2002 6 5 7 8 0
1 2003 6 7 11 8 1
1 2004 12 11 10	9 1
2 2015 6 7 8 9 0
2 2016 3 2 6 3 0
2 2017 13 1 5 6 1
2 2018 7 6 8 12 1
2 2019 5 22 5 6 1
;
run;
data want(drop=i); set have;
	array vars (5) lag4_var lag5_var lag6_var lag7_var treat;
	if treat = 0 then K = lag4_var;
	else do i = 5 to 1 by -1;
		if vars{i} >= 10 then do;
			K = vars{i};
			leave;
		end;
	end;
run;