[R-sig-ME] conditional modes
Gustavo Betini
betinig at uoguelph.ca
Fri Apr 22 11:09:10 CEST 2011
Hi,
I have the following model
mf1<- lmer(y ~ x1 + x2 + (1+x2|ID), data = data)
The conditional modes of the random effects give me a value for the intercept and the slope with respect to x2 for each ID. For example:
ranef(mf1)
ID (Intercept) x2
1 -0.8800 -0.0018
2 -0.7039 -0.0569
3 0.1460 0.0457
...
The conditional modes are standardized values in a sense that negative intercept and slopes are small than the average of the population, right? Lets say that I want to use the conditional modes to investigate if the IDs are decreasing their response with respect to x2. For example, I want to plot my response variable against x2 and have a line for each ID. My concern is that if I use the slopes from the conditional modes, the negative values do not necessarily represent individuals that are decreasing their response with respect to x2. Is that right?
Any help would be much appreciated.
Thanks,
Gustavo Betini, PhD candidate
Dept of Integrative Biology
University of Guelph, Canada
More information about the R-sig-mixed-models
mailing list