Generalized Additive Models
John M. Grego
Abstract
John M. Grego
Abstract
Abstract Generalized additive models replace the linear predictor in a linear model or generalized linear model framework with an additive predictor composed of nonparametric or semiparametric functions of the explanatory variables. Similarly to generalized linear models, the predictor is linear in the link function of the response mean, and the response variable can have a non normal distribution. The last decade has seen interesting new approaches to estimation of generalized additive models, and the development of a mixed effects framework for these models.
OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Abstract Generalized additive models replace the linear predictor in a linear model or generalized linear model framework with an additive predictor composed of nonparametric or semiparametric functions of the explanatory variables. Similarly to generalized linear models, the predictor is linear in the link function of the response mean, and the response variable can have a non normal distribution. The last decade has seen interesting new approaches to estimation of generalized additive models, and the development of a mixed effects framework for these models.
Key concepts: Generalized additive model, Generalized linear model, Generalized linear mixed model, Additive model, Mathematics, Hierarchical generalized linear model, Linear model, Applied mathematics