On Tue, 29 Nov 2016 17:22:47 +0000 (UTC) Elham - via R-help <r-help@r-project.org> wrote:
> Is there another way (I prefer a way with Excel)? Search on "friends don't let friends use excel for statistics." Spreadsheets are an inherently perilous way to do statistics and Excel specifically is notoriously poor. In fact the reason I originally began using dedicated statistical packages (STATA first and now R) is that spreadsheets (Excel in my case) can throw subtle errors that can create problems. immediately, or even worse - later. I had Excel return a negative variance. Since variance is a squared value, unless you are dealing with some very exotic numbers including imaginary values, a negative variance is an absurd result. Further investigation revealed that other stat routines provided with Excel at the time were also throwing errors that could look reasonable and thus be missed. It was also simply using erroneously constructed methods and providing outright wrong results to things like Chi-square calculations. -- John ______________________________________________ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
