Open joelpick opened 3 years ago
Maybe have an additional argument in the list, something like covariate=TRUE, so indicate that the levels of the factor should be taken as a continuous variable. Then could created a crossed time variable in the structure, that could then be used to make various linear/non-linear effects. This could then be used in the sampling stage, do get temporal sampling schemes
everything done apart from temporal autocorrelation
work out how to put temporal effects in