Thanks for a very useful and interesting paper. This seems very relevant in a lot of fields. Since the title contains ".. an introduction with mgcv", I have question about a warning I get when trying to run the random smooth models. Below is an example where I am running Model 2 on page 17:
#> Loading required package: nlme
#> This is mgcv 1.8-27. For overview type 'help("mgcv-package")'.
data <- CO2
data$Plantuo <- factor(as.character(data$Plant))
CO2mod2 <- gam(log(uptake) ~ s(log(conc), k = 5, m = 2) +
s(log(conc), Plant_uo, k = 5, bs = "fs", m = 2),
data = data, method = "REML")
#> Warning in gam.side(sm, X, tol = .Machine$double.eps0.5): model has
#> repeated 1-d smooths of same variable.
As you can see, mgcv generates a warning. I cannot find any mention of such a warning in the paper, so I wonder if this is (a) me doing something wrong, or (b) something that has changed in a new version of mgcv?
In any case, do you consider it safe to ignore this warning?