question about discretionary accrual models
Posted: 27 October 2013 07:43 AM   [ Ignore ]
Sr. Member
Total Posts:  169
Joined  2011-09-20

Dear Joost and the forum,

I read the discretionary accrual sas code again recently, I noticed that the main regression is by cyear sic2. Is there any reason we need to run this model on the calander year?

I think we are using accruals in other research on the fiscal year, such as aduit fee = discretionary accrual + other controls. This kind of model will have both indpendent and dependent variables defined on the fiscal year, am i right?

Based on this, are we suppose to run the discretionary accrual on the fiscal year, instead of cyear?

Any comments are highly appreciated! smile



A humble student of business

Posted: 28 October 2013 06:24 AM   [ Ignore ]   [ # 1 ]
Total Posts:  901
Joined  2011-09-19

hi Zenghui,

Yes, I noticed that too. I am not sure what the idea would be behind this. It probably doesn’t matter smile I will ask the author though.

update: the author (Adam Yore) mentioned it shouldn’t matter much, but the idea was to ‘undo’ any changes in fiscal year (for the few firms that change their fiscal year).

best regards,



To reply/post new questions: Please use the group WRDS/SAS on Google Groups!