library(mgcViz)
set.seed(2) ## simulate some data...
dat <- gamSim(2,n=500,dist="normal",scale=0.25)$data
# Fit GAM and get gamViz object
b <- mqgamV(y~s(x) + s(z) + I(x*z), data = dat, qu = c(0.25, 0.5, 0.75),
aQgam = list(argGam = list(select = TRUE)), aViz = list("nsim" = 0))
# This is equivalent to doing
# 1. Fit QGAM
# b <- mqgam(y~s(x) + s(z) + I(x*z), data=dat,
# qu = c(0.25, 0.5, 0.75), argGam = list(select = TRUE))
# 2. Convert to gamViz object
# b <- getViz(b, nsim = 0)
# Either way, we all effects by doing
print(plot(b, allTerms = TRUE), pages = 1)
Run the code above in your browser using DataLab