How to use SAS/SQL to create a table with certain conditions from a dataset - sql

Question

I have a dataset with ID and event_year (event meaning something happened that year, a person has more than one record in this table with more than one event year eg. ID 1 can have three entries with event_year 2017, 2018, 2019 ). Example dataset like:
ID event_year
1 2017
1 2018
1 2019
2 2018
2 2017
ID
event_year
1
2017
1
2018
1
2019
2
2018
2
2017
I need to get a table from this of all ID where the event_year is between 2017 and 2021 to make a frequency table counting people with event_year at set years 2017, 2018, 2019, 2020, 2021 (these are the columns refer to as study year x).
Year frequency
2017 2
2018 2
2019 1
2020 1
2021 0
Year
frequency
2017
2
2018
2
2019
1
2020
1
2021
0
Another condition is for the study year x if a person didnt have an event_year in x but had event_year x-1 they will be included in the frequency of year x, for example the ID 1 above should be included in frequency of once in each 2017, 2018, 2019 and 2020- because following the condition above for year 2020 they didnt have event_year in 2020 but did in 2019 so will be included in 2020. I apologise if this is confusing and would be happy to clarify

Answer 1

If I understood your question, this should work:
data have;
input ID event_year;
datalines;
1 2017
1 2018
1 2019
2 2018
2 2017
3 2017
3 2020
;
run;
For the next step (your additional requirement of being included a year after last event) we need data grouped by ID.
proc sort data=have;
by ID;
run;
We just add extra rows to a table, where a year is last year + 1.
data have;
set have;
output;
by ID;
if last.ID then do;
ID=ID;
event_year=event_year+1;
output;
end;
run;
Now we just check how many different IDs every year had. If you want to check only for certain years, just add a where clause (for example, where event_year in (2017, 2018, 2019, 2020, 2021) ).
proc sql;
create table want as
select distinct event_year, count(distinct ID) as frequency
from have
group by event_year
;
run;