I know this was asked and answered in 2005, I just want to make sure the answer still holds in 2010. Question: If I have the mixed-model with lmer( y ~ x1 + x2 + (1 | id) + (z1 | w ) , ... ) there is no way for me to specify a correlation structure for the random effects? Right? If I want to specify correlation structure then I need to use lme in the nlme package? Right? Crossed Random effects are coded with the pdBlocked syntax? Right? Thanks, -tgs [[alternative HTML version deleted]]
Thomas Stewart <tgstewart <at> gmail.com> writes:> If I have the mixed-model with > > lmer( y ~ x1 + x2 + (1 | id) + (z1 | w ) , ... ) > > there is no way for me to specify a correlation structure for the random > effects? Right? > > If I want to specify correlation structure then I need to use lme in the > nlme package? Right? Crossed Random effects are coded with the pdBlocked > syntax? Right?Right on all counts (see Pinheiro and Bates 2000 for details of crossed-effect coding with pdBlocked). In general it might be better to send these types of questions to r-sig-mixed-models instead of r-help ...