Dear list,
I'm trying to calculate a regression on a dataset with repeated measures. I
tried to look for an example on the web and in Pinheiro and Bates
"Mixed-Effects Models in .. " book.
However, I' m not sure wether the regression model I'm using is the
"right" one. I'm very thankful for any suggestions!
Ok, here's what I'm trying to do:
First, the experimental design:
We ran a study, where 36 subjects played a game, where they could 150 times make
risky decisions. Throughout the game we recorded physiological data.
Thus, each subject made in every trial a more or less risky decision. Now,
I'm interested, whether parameters of the physiological data (Phys1, Phys2)
can predict the amount of risk taken by a subject.
So far my regression model looks as follows:
lme(Risk ~ Phys1 + Phys2, random = ~ 1 | Subject)
Could this be suitable? Is it a problem, if the Variable Risk is autocorrelated
within Subjects?
Thank you very much for any help and sorry if this is a too trivial question for
the list.
Kind regards
Andreas Pedroni
