Timeline for Replicate t or F test from regression using regression likelihoods
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 8 at 21:42 | answer | added | Sextus Empiricus | timeline score: 4 | |
| Apr 8 at 7:33 | comment | added | Christoph Hanck | See stats.stackexchange.com/questions/449344/… and stats.stackexchange.com/questions/276192/… | |
| Apr 8 at 0:21 | comment | added | A Friendly Fish | can i cite you and your answer in my publication? i don't think it's a problem to cite you as whuber, but if you prefer another name, please let me know. :) thanks! | |
| Apr 7 at 16:05 | answer | added | Zhanxiong | timeline score: 6 | |
| Apr 7 at 15:10 | comment | added | whuber♦ | Unless the errors are iid Normal, both p-values are approximations: different ones. Also, the distribution of the likelihood ratio is asymptotic. You should be pleased to get such close agreement with only 48 df in any single dataset. If you wish to compare the two statistics, then please simulate your data and do so repeatedly so you can study how the differences vary. | |
| Apr 7 at 1:59 | history | asked | A Friendly Fish | CC BY-SA 4.0 |