Chapter 8 GAM validation | Workshop 8: Generalized additive models in R - QCBS So, what can we do if the observations of the response variable do not follow a Normal distribution? Or if the variance is not constant (heteroscedasticity)? Just like generalized linear models (GLM), we can formulate generalized additive models to deal with these issues
How to choose family in Generalized Additive Model (GAM) When modelling a GAM model using mgcv in R, we need to define the family = I tried some families (e g , Gaussian, Gamma), R seems to build them all successfully Are there some guidelines about how to choose the appropriate "family"? You have to think about the distribution of the outcome conditioned on the covariates
Checking, Selecting Predicting with GAMs - School of Mathematics Since a GAM is just a penalized GLM, residual plots should be checked, exactly as for a GLM The distribution of scaled residuals should be examined, marginally, and plotted against covariates and fitted values residuals(model) extracts residuals gam check(model) produces simple residual plots, and summary ̧ estimation convergence information
normal distribution - GAM with opposite outcomes with different families - Cross Validated From what I've read, looking at the histogram above, the distribution that would fit best is the gamma distribution, or eventually the inverse gaussian Also the residuals look better: s(month, by=factor(s_status)) + s(factor(ds_2$id), bs="re"), data=ds_2, family = Gamma(link = "inverse"), method = "REML")