Generalized Additive Model (GAM)
Generalized additive model (GAM) is a term that is used in statistics as a means of analyzing data derived from research. It is also called “wiggly model” and is a nonlinear model method used in regression analysis, the statistical method of examining or estimating the relationship among variables. GAMs were first developed by Trevor Hastie and Robert Tibshirani mixing properties of generalized linear models (GLMs) and additive models. There are a lot of data in the environmental sciences that do not fit simple linear models and are best described by these “wiggly models”. GAMs allow a researcher to incorporate nonlinear and other relationships into otherwise linear models.
Generalized linear models (GLMs) are used to predict a response (dependent) variable from one or more predictor (independent) variables when the response data does not follow a normal distribution model. Additive models are used when the data being subjected to regression analysis is non-parametric, i.e. categorical and ranked (nominal and ordinal data).