Generalized Additive Model

Interpretable model composed of univariate and bivariate shape functions for regression

Use fitrgam to fit a generalized additive model for regression.

A generalized additive model (GAM) is an interpretable model that explains a response variable using a sum of univariate and bivariate shape functions of predictors. fitrgam uses a boosted tree as a shape function for each predictor and, optionally, each pair of predictors; therefore, the function can capture a nonlinear relation between a predictor and the response variable. Because contributions of individual shape functions to the prediction (response value) are well separated, the model is easy to interpret.

Objects

`RegressionGAM`	Generalized additive model (GAM) for regression (Since R2021a)
`CompactRegressionGAM`	Compact generalized additive model (GAM) for regression (Since R2021a)
`RegressionPartitionedGAM`	Cross-validated generalized additive model (GAM) for regression (Since R2021a)
`RegressionChainEnsemble`	Multiresponse regression model (Since R2024b)
`CompactRegressionChainEnsemble`	Compact multiresponse regression model (Since R2024b)

Functions

expand all

Create GAM Object

`fitrgam`	Fit generalized additive model (GAM) for regression (Since R2021a)
`compact`	Reduce size of machine learning model
`crossval`	Cross-validate machine learning model
`templateGAM`	Generalized additive model (GAM) learner template (Since R2023b)

Update GAM

`addInteractions`	Add interaction terms to univariate generalized additive model (GAM) (Since R2021a)
`resume`	Resume training of generalized additive model (GAM) (Since R2021a)

Interpret Prediction

`lime`	Local interpretable model-agnostic explanations (LIME) (Since R2020b)
`partialDependence`	Compute partial dependence (Since R2020b)
`permutationImportance`	Predictor importance by permutation (Since R2024a)
`plotLocalEffects`	Plot local effects of terms in generalized additive model (GAM) (Since R2021a)
`plotPartialDependence`	Create partial dependence plot (PDP) and individual conditional expectation (ICE) plots
`shapley`	Shapley values (Since R2021a)

Assess Predictive Performance on New Observations

`predict`	Predict responses using generalized additive model (GAM) (Since R2021a)
`loss`	Regression loss for generalized additive model (GAM) (Since R2021a)

Assess Predictive Performance on Training Data

`resubPredict`	Predict responses for training data using trained regression model
`resubLoss`	Resubstitution regression loss

Assess Predictive Performance on Cross-Validated Data

`kfoldPredict`	Predict responses for observations in cross-validated regression model
`kfoldLoss`	Loss for cross-validated partitioned regression model
`kfoldfun`	Cross-validate function for regression

Use Regression Chains for Multiresponse Regression

`fitrchains`	Multiresponse regression with regression chains (Since R2024b)
`compact`	Reduce size of multiresponse regression model (Since R2024b)
`loss`	Loss for multiresponse regression model (Since R2024b)
`predict`	Predict responses using multiresponse regression model (Since R2024b)

Topics

Train Generalized Additive Model for Regression
Train a generalized additive model (GAM) with optimal parameters, assess predictive performance, and interpret the trained model.