Require non-missing values of control variables

[Jet Struik]

I am performing a replication study in which I replicate a study of Dyreng et al. (2017). One of the requirements on the dataset used by Dyreng et al. is "Require non-missing values of control variables".
At first, I tried to use the command drop if missing(variable). However, this reduced my sample with over 95% and this is not the intention, so I assume this is not what is meant by the requirement.
What is meant by this requirement and how should this be implemented in the data in Stata?
Thanks in advance!

[Carlo Lazzaro]

Jet:
I do not know the paper that you mention (BTW: as per FAQ, full reference, please).
It might be that they performed a complete case analysis, just leaving in all the observations with observed values for all the variables under investigation.
That said, as Stata uses matrices behind the scene, it calls listwise deletion by default, ruling out form the subsequent statistical procedure all the observations with at least one missing values in the variables on interest.

[Jet Struik]

Thanks for your reply!
I understand what you mean. Full reference of the paper is: Dyreng, S. D., Hanlon, M., Maydew, E. L., & Thornock, J. R. (2017). Changes in corporate effective tax rates over the past 25 years. Journal of Financial Economics, 124(3), 441–463. https://doi.org/10.1016/j.jfineco.2017.04.001
In their paper, they present in the sample composition table 1 on page 5 a reduction in the firm-years and firms (the observations) after the requirement of non-missing control variables. That's why I think there must be some way in Stata to implement this requirement into the data, the same way as the other requirements can be. Those I did already like this:

[Carlo Lazzaro]

Jet:
thanks for providing full reference of the paper.
I would repeat step by step the reduction detailed in the paper (assuming that you have the very same dataset) and check whether the sample size reaches "4643 unique firms, for an average of 11.6 years of data per firm".