[R] How to prevent inclusion of intercept in lme with interaction

[Christian Ritz]

Hi Dieter,

the following model assumes a linear relationship between the response "newbone" and the
independent variable "t" with a common intercept equal to 0 and treatment-dependent slopes:


grd.lme0 <- lme(newbone~t:treat-1, data=grd, random=~1|subject)

summary(grd.lme0)




Christian