Dear all:
I m a newer on R.? I have some problem when I use?anova function.? I use anova function to get Type I SS results, but I also need to get Type III SS results.? However, in my code, there is some different between the result of Type I SS and Type III SS.? I don?t know why the ?seqe? factor disappeared in the result of Type III SS.? How can I do??
Here is my example and result.
a<-c(1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13,14,14)
b<-c(1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2)
c<-c(2,1,1,2,2,1,1,2,2,1,1,2,2,1,1,2,2,1,1,2,2,1,1,2,2,1,1,2)
d<-c(2,2,1,1,2,2,1,1,2,2,1,1,2,2,1,1,2,2,1,1,2,2,1,1,2,2,1,1)
e<-c(1739,1633,1481,1837,1780,2073,1374,1629,1555,1385,1756,1522,1566,1643,
???? 1939,1615,1475,1759,1388,1483,1127,1682,1542,1247,1235,1605,1598,1718 )
KK<-data.frame(subj=as.factor(a), drug=as.factor(b), per=as.factor(c),? seqe=as.factor(d), Cmax=e)
M<- lm(Cmax ~ seqe+ subj:seqe + per + drug , data=KK)
anova(M)
drop1(M, test="F")
The result of Type I SS:
Analysis of Variance Table
Response: Cmax
????????? Df Sum Sq Mean Sq F value Pr(>F)
seqe?????? 1??? 585???? 585? 0.0160 0.9014
per??????? 1? 63175?? 63175? 1.7293 0.2131
drug?????? 1? 58149?? 58149? 1.5917 0.2311
seqe:subj 12 634325?? 52860? 1.4469 0.2660
Residuals 12 438395?? 36533
The result of Type III SS:
Single term deletions
Model:
AUCt ~ seqe + subj:seqe + per + drug
????????? Df Sum of Sq????? RSS????? AIC F value? Pr(F)
<none>???????????????? 63208187????? 442
per??????? 1?? 2100484 65308672????? 441? 0.3988 0.5396
drug?????? 1?? 4714183 67922370????? 442? 0.8950 0.3628
seqe:subj 12? 35813062 99021249????? 430? 0.5666 0.8308
Best regards,
HY Lee
_________________________________________________________________________________________________________
?????????????? http://tw.promote.mail.yahoo.com/dc/change.html