[Gretl-users] Gretl: means by categorical variable

Hi Allin Thanks for responding. "What I think we should do is make a special case of original data that take the form of very large integers (e.g. population, and not many other things), and show corresponding large integers for mean, median, minimum and maximum." If gretl could show corresponding large integers, that would be great. Thanks Gene On Monday, May 27, 2024 at 08:59:06 PM EDT, Allin Cottrell <cottrell(a)wfu.edu&gt; wrote: On Mon, 27 May 2024, g s wrote: It wouldn't be a great deal of work, but -- sorry -- it would be crazy. Showing 13 significant digits would be _way_ beyond the accuracy of the original data, and doing this for several of the statistics shown by the "summary" command would push the line length way beyond what's reasonable. What I think we should do is make a special case of original data that take the form of very large integers (e.g. population, and not many other things), and show corresponding large integers for mean, median, minimum and maximum. So, for example, 1557590 for the Australia-Oceania mean population. I take you mean, show the sum. That's not a statistic that's often wanted, and it's very easy to compute if you do want it, so I think probably not. Allin _______________________________________________ Gretl-users mailing list -- gretl-users(a)gretlml.univpm.it To unsubscribe send an email to gretl-users-leave(a)gretlml.univpm.it Website: https://gretlml.univpm.it/postorius/lists/gretl-users.gretlml.univpm.it/