COmparison between rows of a dataset

[ArchanaSudhir]

HI I have a dataset like this

sort1	sort2	var1	var2	var3	var4	var5	var6
1	1	5	0
2	2	1 2 3 4	7 8 9 5	0.1	0.1	0.1	2
3	2	1 2 3 5	3 4 5 6	0.1	0.1	0.1

If var2 is 0 in the first line ,

I need to set var3,var4,var5,var6 as zero in teh next 2 lines.

How do I do this?

Thanks,

Archana

[andy_smith_amadeus_co_uk]

Hi Archana,

This looks like a job for the retain statement to set up a flag.

How about something like:

data new;

  retain delflag; * don't reset this value to missing for every iteration;

  set old;

  if _n_ eq 1 and var2=0 then delflag="Y"; * check for var2=0 only on the first iteration;

  if _n_ gt 1 and delflag="Y" then do;

    var3=0;

    var4=0;

    var5=0;

    var6=0; * Set to zeros only if correct condition met;

  end;

run;

[PGStats · Accepted solution]

A sum statement creates a retained variable which is useful here

data a;
infile datalines truncover;
input sort1	sort2	var1	var2	var3	var4	var5	var6;
datalines;
1	1	5	0	 	 	 	 
2	2	1 2 3 4	7 8 9 5	0.1	0.1	0.1	2
3	2	1 2 3 5	3 4 5 6	0.1	0.1	0.1	
;

data want;
set a;
if var2 = 0 then doit = 2;
else if doit then do;
    var3=0; var4=0; var5=0; var6=0;
    doit + (-1);
    end;
drop doit;
run;

[Astounding]

It gets a little more complex if VAR2 could be 0 on consecutive observations.

[LinusH]

Your requirement together with the record layout gives me the feeling that you don't store the data in an optimal way.
Can you tell us the reason for this logic, and what var2 and var3-var6 represents in the real world?