Hello, I have a data set with about 5 million observations. This table contains an index with several columns, including one column for the year and one column for the country. I want to delete all observations of a certain country for a certain year from this table. if country = 'NZ' & year = 2018 then delete; Is there an efficient way to do this and make use of the indexed data? So far I have tried the following two solutions, which have a too long runtime of about 30 minutes (both). proc sql;
delete from mydata
where country = 'NZ' & year = 2018;
quit;
data mydata;
modify mydata;
if country = 'NZ' & year = 2018 then remove;
run; In case it matters: the table is sorted by the country and year columns. I want to avoid writing a data set where the index is simply recreated. But perhaps it is the most efficient solution? Thanks for your help.
... View more
