Re: [R] How to test combined effects?

Ooops. One typo in the estimable command: estimable(ModelFit, c('IQ:age'=1, 'IQ:I(age^2)'= 1, 'IQ:I(age^3)' = 1)) (Remove a trailing space in the second string.) -G On Nov 2, 2007, at 11:51AM , Gregory Warnes wrote: Hello Gang, First, if you would like to performa an overall test of

Re: [R] How to test combined effects?

Thanks a lot for the help! First, if you would like to performa an overall test of whether the IQ interactions are necessary, you may find it most useful to use anova to compare a full and reduced model. Something like: ModelFit.full -lme(mct~ IQ*age+IQ*I(age^2)+IQ*I(age^3),

Re: [R] How to test combined effects?

Gang Chen gangchen at mail.nih.gov writes: Suppose I have a mixed-effects model where yij is the jth sample for the ith subject: yij= beta0 + beta1(age) + beta2(age^2) + beta3(age^3) + beta4(IQ) + beta5(IQ^2) + beta6(age*IQ) + beta7(age^2*IQ) + beta8(age^3 *IQ) +random

Re: [R] How to test combined effects?

Dieter, Thank you very much for the help! I tried both glht() in multcomp and estimable() in gmodels, but couldn't get them work as shown below. Basically I have trouble specifying those continuous variables. Any suggestions? Also it seems both glht() and estimable() would give multiple t