After testing, I found that SAS does not make errors. Thank you very much for your contribution. ... View more

@mkeintz Thank you for your assistance.   I have a question. If the company with a big problem is 2 or 42, how can I avoid 2 by catching 2 instead of catching 12 and 42? ... View more

@PaigeMiller Amazing, the code worked. Thank you very much. ... View more

Thanks to the experts for pointing out my problem.   I have revised the data set (experience). It should include the following board work experience. If his work at that time was related to IT profession, then it is 1 (IT_PROFESSION).   The final data set (want), is that I want to identify whether a particular person had IT work experience before serving as a board member that year (IT_EXPERTISE = 1). ... View more

Hello, expert:   The first data set mainly records all the work records of a specific person (including work years and work experience). The second data set is that the specific person was a member of the board of directors in that year.   I want to know whether the particular person had relevant work experience before becoming a board member, which is why the second data set will have more recent years.   Thanks. ... View more

Hello experts @Tom @Patrick @FreelanceReinh ,   I have solved the problem, and the program works successfully. Thank you again. ... View more

Hello expert @Tom,   I need to identify the company announcement date that falls between the start and end dates of the director’s employment, but the end date may display C and N at the same time (as shown in the previous picture).   In my dataset, it shows: C = Current: C means that the director is still in office until now (which is informative).  N = n/a: N means missing values ​​(which have no informational meaning).   If the program is used to allow missing, it may be inappropriate. Thus, I think a more appropriate way is to turn C into a long time in the future and treat N as a missing processing method.   Thank you very much for your comment to make the program more perfect. Respect! ... View more

Hello @FreelanceReinh,  It worked, thank you.   I learned a lot from you, thank you experts @Tom @Patrick @FreelanceReinh ! ... View more

Big thanks to the two experts @Tom and @Patrick .   I may not be completely clear about the format of SAS, but I generally understand that what the experts mean is to create a new format, and then apply this format when the data is newly created, so that C becomes a future date.   However, my dataset is downloaded from the database and imported into SAS in the sas7bdat file format. I have just tried for a long time and it seems that there is no way to convert C into a future date. Maybe I'll just use missing to handle it.       I feel like two experts did their best, thank you very much anyway. ... View more

First, I would like to thank the two experts (@Patrick and @Tom) for your assistance.   I have a question that I would like to ask you further. The director's start and end information I obtained from the database shows:    C = Current: C means that the director is still in office until now (which is informative). Although it is displayed in text, the format is still a date (as shown in the picture). N = n/a: N means missing values ​​(which have no informational meaning).     I think a more appropriate way is to turn C into a date today or a long time in the future, and treat N as a missing processing method. Do you know how to adjust the program? Thanks again. ... View more

Sorry to bother you again.   In the data I downloaded, if the directors have been in the current company up to now, the date column shows C representing "current".   data experience; input PERSON $ COMPANY_ID (START_DATE END_DATE) (:yymmdd8.) ind $; format START_DATE END_DATE yymmddn8. ; cards; A 101 20050101 20121231 YES A 103 20070601 20241031 NO B 109 20140503 C YES B 110 20180820 20221225 YES ; run;     This causes the following syntax to fail to execute. Do you know of any solution?   l.announce_date between r.start_date and r.end_date     ... View more

It worked, thank you. ... View more

Hi expert, I would like to use existing experience data and identify whether the director is working in the company and whether he is independent. Those numbers are dates, so I changed the format of the template.   data experience; input PERSON $ COMPANY_ID START_DATE END_DATE IND $; format START_DATE yymmddn8. ; /*The date on which the director started working for the company*/ format END_DATE yymmddn8. ; /*The date on which the director ended working for the company*/ cards; A 101 20050101 20121231 YES A 103 20070601 20241031 NO B 109 20140503 20220429 YES B 110 20180820 20221225 YES ; run; data have; input PERSON $ COMPANY_ID ANNOUNCE_DATE ; format ANNOUNCE_DATE yymmddn8. ; /*The company publicly announces that the director is serving the company*/ cards; A 103 20071231 B 109 20151231 ; run; data want; input PERSON $ COMPANY_ID ANNOUNCE_DATE IND $; format ANNOUNCE_DATE yymmddn8. ; cards; A 103 20071231 NO B 109 20151231 YES ; run; /*If the date announced by the company is between the start and end dates of work experience, the director’s independence can be identified*/   ... View more