Timeline for How to define contrast coefficient matrix?
Current License: CC BY-SA 3.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 13, 2017 at 12:44 | history | edited | CommunityBot |
replaced http://stats.stackexchange.com/ with https://stats.stackexchange.com/
|
|
| Oct 22, 2012 at 5:35 | comment | added | user1172558 | Yes. x1 and x2 are factors, but not pre. I posted the str(data) above. | |
| Oct 22, 2012 at 5:30 | comment | added | mnel |
Are you sure that x1 and x2 are coded as factors? Should pre be a factor?
|
|
| Oct 22, 2012 at 5:29 | review | First posts | |||
| Oct 22, 2012 at 5:52 | |||||
| Oct 22, 2012 at 5:27 | comment | added | user1172558 | I don't know why, but I can't get this result. Even using drop1() options(contrasts=c("contr.sum","contr.poly")) m2 <- lm(y ~ pre + x1*x2, data) summary(m2) anova(m2) drop1(m2, .~., test='F') # Type III SS The only result which is still correct is the one for 'pre'. | |
| Oct 22, 2012 at 5:21 | vote | accept | user1172558 | ||
| Oct 22, 2012 at 5:22 | |||||
| Oct 22, 2012 at 5:15 | history | edited | mnel | CC BY-SA 3.0 |
added 63 characters in body
|
| Oct 22, 2012 at 5:10 | history | answered | mnel | CC BY-SA 3.0 |