There is absolutely nothing at the moment about REML vs ML. A broader issue is that we need some explanation for how the parameters are estimated. This is explained for the simple linear model in ch 5, but not for the LMM, even though this is not so hard if one know matrix algebra at an elementary level.
There is absolutely nothing at the moment about REML vs ML. A broader issue is that we need some explanation for how the parameters are estimated. This is explained for the simple linear model in ch 5, but not for the LMM, even though this is not so hard if one know matrix algebra at an elementary level.