This is an internal function contained in the multiFAMM function. This step uses the information from the univariate FLMMs for the MFPCA. It also allows a simple weighting scheme of the MFPCA.
prepare_mfpca(model_list, fRI_B, mfpc_weight)
Arguments
model_list: List containing sparseFLMM objects for each dimension as given by the output of apply_sparseFLMM()
fRI_B: Boolean for including functional random intercept for individual (B in Cederbaum). Defaults to FALSE.
mfpc_weight: TRUE if the estimated univariate error variance is to be used as weights in the scalar product of the MFPCA.