I have a question regarding the filtering of a loan dataset for my upcoming thesis.
My dataset consists of loan data which is reported for 5 years on a quarterly basis. The column of interest is the 'Loan Identifier' as well as the 'Cut-Off-Date'. I just want to observe the loans (via Loan Identifier) that exist at the first reporting date (first quarter) for every upcoming quarter (cut-off-date).
For example, if there are the loans with the identifier c("1001","1002","1003") in the first cut-off-date and the second cut-off date, one quarter later, has loans with identifiers ("1002","1003","1004"), R should filter for only the identifiers that existed in the first quarter ("1002","1003"). So that new loans during the analysis are completely ignored.
Is there also the possibility to do that all in one file? Or should I extract the data of each cut-off-date in a new table?
Thanks and best regards!
I am thinking about assigning each loan in the first quarter as a vector. After that, I should split up the loan dataset for each cut-off-date and merge the vector with the new tables via left_join. So that every loan that does not match with the vector is disregarded.
As I have multiple loan pools with 15 pool-cut-off dates, this seems very impractical for me. Maybe there is a smarter and more effective solution.
