Entering edit mode
Jianping Jin
▴
890
@jianping-jin-1212
Last seen 10.2 years ago
Dear statisticians of the list,
I am a limma user and have a small data set that I used the lme mixed
effect model for analysis. I need some help in extracting comparisons
from the result. I sent my questions to the R help list, but it
appears that there is no people there care about the questions. I know
my questions may not be related to this list, but if you would help
could you please point me out what I can do or I should do something
else? Your help will be greatly appreciated.
My data set is similar to the example data, Orthodont, and I used
patient subject as a random effect and did something below:
library(nlme)
library(gregmisc)
data(Orthodont)
Orthodont$AgeGroup <- gtools::quantcut(Orthodont$age)
fm <- lme(distance ~ AgeGroup + Sex + AgeGroup*Sex, data = Orthodont,
random = ~1|Subject)
fit.contrast(fm, "Sex", c(-1,1), conf=0.95)
fit.contrast(fm, "AgeGroup", rbind("1 vs 2"=c(-1,1,0,0), "2 vs
3"=c(0,-1,1,0)), conf=0.95)
But how can I compare different age groups among Male, Female or
between Male and Female as illustrated below (age groups on rows):
## comparison among Male
Male Female
[1,] -1 0
[2,] 1 0
[3,] 0 0
[4,] 0 0
## among Female
Male Female
[1,] 0 0
[2,] 0 -1
[3,] 0 0
[4,] 0 1
## between Male and Female
Male Female
[1,] -1 0
[2,] 0 0
[3,] 0 0
[4,] 0 1
JP Jin Ph.D.
Bioinformatics scientist
Center for Bioinformatics
UNC at Chapel Hill
Chapel Hill, NC
[[alternative HTML version deleted]]
