heterogeneous error structure

Hello all, I am attempting to fit a random regression model using orthogonal polynomials of time to model additive effects on a trait in a wild sheep population and was wondering if anyone could help me with coding the R structure. I can fit this model with an assumed homogeneous (with time) error variance, or with a multivariate error structure (dividing the time period into discrete age classes). However, I've read a few papers discussing the assumption of a structural model on the residual variance (e.g. as a quadratic or other polynomial function), such that Vr is a continuous function of time. I'd like to try this. My question is, can anyone tell me how to code the R structure to do this? Unfortunately all my semi-intuitive guesses are yielding a " Failed to parse R structure line" error message. Below I have appended the model I have been able to fit with the homogenous error variance (as well as an additive effect as a quadratic, a PE and a maternal genetic effect). Any help or suggestions would be greatly appreciated. Alastair J Wilson PhD # AGEF age as a factor, sAGE as continuous standardized -1 to1 TRAIT ~ mu SEX AGEF !r pol(sAGE,2).ANIMAL ide(ANIMAL) MOTHER !f mv 1 1 3 0 0 ID 3 # single homogeneous error variance pol(sAGE,2).ANIMAL 2 pol(sAGE,2) 0 US 1.64 0.96 7.801 0.01 0.01 0.01 ANIMAL ide(ANIMAL) 1 ide(ANIMAL) 0 IDV 4.55 MOTHER 1 MOTHER 0 IDV 0.226 Alastair.Wilson_at_ed.ac.uk tel. 0131 6507287 Institute of Evolutionary Biology School of Biological Sciences The University of Edinburgh Ashworth Laboratories The King's Buildings West Mains Road Edinburgh EH9 3JT, UK
Received on Mon Mar 08 2005 - 12:40:47 EST