This function combines the robustgam with automatic smoothing parameter selection. The smoothing parameter is selected through robust cross validation criterion described in Wong, Yao and Lee (2013). The criterion is designed to be robust to outliers. This function uses grid search to find the smoothing parameter that minimizes the criterion.
This function is the same as robustgam.GIC, except that the R internal optimization function optim is used to find the smoothing parameter that minimizes the RAIC or RBIC criterion.
This function combines the robustgam with automatic smoothing parameter selection. The smoothing parameter is selected through generalized information criterion (GIC) described in Wong, Yao and Lee (2013). The GIC is designed to be robust to outliers. There are two criteria for user to choose from: Robust AIC and Robust BIC. The robust BIC usually gives smoother surface than robust AIC. This function uses grid search to find the smoothing parameter that minimizes the criterion.
This package provides robust estimation for generalized additive models. It implements a fast and stable algorithm in Wong, Yao and Lee (2013). The implementation also contains three automatic selection methods for smoothing parameter. They are designed to be robust to outliers. For more details, see Wong, Yao and Lee (2013).
● Data Source:
CranContrib
● Keywords: Bounded score function, Generalized information criterion, Generalized linear model, Robust estimating equation, Robust quasi-likelihood, Smoothing parameter selection
● Alias: robustgam-package
●
0 images
This function implements a fast and stable algorithm developed in Wong, Yao and Lee (2013) for robust estimation of generalized additive models. Currently, this implementation only covers binomial and poisson distributions. This function does not choose smoothing parameter by itself and the user has to specify it manunally. For implementation with automatic smoothing parameter selection, see robustgam.GIC, robustgam.GIC.optim, robustgam.CV. For prediction, see pred.robustgam.