Hi,
My collegue has asked me to calculate the steady state of a pharmacokinetic
study.
This is defined as where the concentration after a certain time doesn't
increase anymore.
So if I perform multiple t-tests between the sample points, with alternative
less, then I can see which from which time point there isn't an increase in
concentration. But this doesn't feel good to me. I don't think that this
is the right way to check if the concentration is still increasing. And if I
perform multiple comparisons, I should use the bonferroni correction?
I was thinking about an ANOVA test to find differences between the time points,
but as there are no replications and it's a paired t-test, I'm not able
to calculate an ANOVA.
Below you find the data, I hope someone can point me to the right way.
Kind regards
Bart
dat <-structure(list(Sample = 1:8, X12 = c(0.305, 0.44, 0.6, 0.47, 0.49,
0.17, 0.435, 0.435), X36 = c(0.84, 1.16, 1.32, 1.17, 0.78, 0.43, 0.93, 0.9), X60
= c(1.05, 0.9, 0.98, 1.03, 0.77, 0.47, 0.85, 1.11), X84 = c(0.87, 0.95, 1.31,
1.13, 0.62, 0.93, 1.02, 0.79), X108 = c(0.79, 0.92, 1.29, 0.97, 0.74, 0.86,
1.03, 1.41)), .Names = c("Sample", "X12", "X36",
"X60", "X84", "X108"), class =
"data.frame", row.names = c(NA, -8L))
for (i in 2:5) { print(paste((names(dat)[c(i,i+1)]),collapse="-"))
print(t.test(x=dat[,i],y=dat[,i+1],alternative="less",paired=T,conf.level=0.95))}
_________________________________________________________________
[[alternative HTML version deleted]]
