[R] Linear mixed-effect models and model selection

David_txertudi at yahoo.com David_txertudi at yahoo.com
Tue Jul 24 08:51:51 CEST 2012

Not sure about the first question without knowing more about your model and research aims. As for the second, there are a number of methods that can be used to select models--F tests and other variance comparisons being among the most common.  Given your examiner's comment about parsimony, I'm thinking a BIC or AIC might be helpful. Available in R as AIC() and BIC(), they describe the tradeoff between model complexity and accuracy (more or less). Both penalize additional terms (BIC does so more strongly) to avoid "over fitting", but reward goodness of fit in likeliness models. They do not by themselves provide an absolute measure of goodness of fit, however.   Does that help at all?


On Jul 23, 2012, at 7:22 PM, fariba moslih <shakhenabat1 at yahoo.com.au> wrote:

> Hi,
> I am looking at the effect of allelochemicals produced by two freshwater macrophyte species on two different algal species at different days. I am comparing the effect of each macrophyte on each algae at each day. I received help from someone doing the LMEM (Linear mixed-effect models) and he told me to do ANOVA to analyse the LMEM. However, I received these feedback from my examinor;
> 1. An Analysis of Variance (ANOVA) was used to analyse the LMEM. 
> Comment; Not clear why you used ANOVA here. Was it to compare models?
> 2.The effects of TREATMENT and TIME and their interaction were all significant (Table 5).  Because of the significant interaction, the analysis was split by TIME. 
> Comment; Given that you have interactions, you should do a model selection to show whether the interaction model is  in fact more parsimonious. 
> Can someone explain these and tell me how and when should I do model selection? 
> Thanks,
> F
>    [[alternative HTML version deleted]]
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

More information about the R-help mailing list