Timeline for 'one-way anova' randomization test in R
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 9, 2020 at 12:40 | comment | added | antifrax | The relationship between the F statistic and Ernst's "T-statistic" is confusing (he calls them 'equivalent'). This may be because Ernst's statistic follows an F-distribution (as per your histograms above!), but it's not a ratio of variances. | |
| Jul 23, 2020 at 8:13 | comment | added | StupidWolf | Hi @antifrax, there was more than one question and I tried to answer one of them. You can post your answer, most likely I don't have the time to reformulate my answer. From your comment, i gather there's nothing fundamentally wrong. | |
| Jul 23, 2020 at 1:08 | comment | added | antifrax | Thanks @StupidWolf. That's cleaner script! I've looked into the coin package which does basically this (the difference-in-means test) in a more general framework (and I think with some intelligent corrections). I'll post that approach ASAP. I appreciate your answer but I don't think I can "accept" it since it is more of a comment / elucidation than a clear response. Perhaps reformulate it ? | |
| Jun 27, 2020 at 20:17 | history | answered | StupidWolf | CC BY-SA 4.0 |