Dear Forum,

the use of Compustat data in research often varies extensively in terms of the period observed. Sometimes the data that is used goes back to 1976, but very often the early years are excluded and samples are restricted to periods after e.g. 1980. Thus, I would like to know whether there are one or two good sources discussing the reliability of Compustat annual fundamental data in general. I am interested in the reasons why early observations are often excluded and how to justify that in a thesis.

Reasons that came to my mind are missing values, regulatory or administrative issues. However, this is just a guess.

Thanks for your comments and hints.
Chris

hi Chris,

That is indeed an important question; I have not come across such discussions. Feel free to post anything you may find on this topic in this thread.

For your thesis I would think it is ‘ok’ to see how far back the data goes, or to justify based on changes in regulation. For example ‘regulation FD’ (for a disclosure study), or a year where a relevant (for the study) accounting standard changed.

best regards,

Joost