example_ANOVA.R

data <- read.csv("depression.csv")
model <- 'depression ~ factor(group) -1'
R1 <- rbind(c(-1,1,0), c(0,-1,1))

## classical F test ##
fit.lm <- lm(model, data=data)
f <- summary(fit.lm)$fstatistic
# F statistic
f[1]
# p value
pf(f[1],f[2],f[3], lower.tail = FALSE)


## F-bar test ##
fit.csi <- csi.lm(model, data, ui = R1)
# F-bar statistics
fit.csi$T.obs
# p values
fit.csi$p.value