scam provides functions for generalized additive modelling under shape constraints on the component functions of the linear predictor of the GAM. Models can contain multiple shape constrained and unconstrained terms as well as bivariate smooths with double or single monotonicity. The model set up is the same as in mgcv(gam) with the added shape constrained smooths, so the unconstrained smooths can be of more than one variable, and other user defined smooths can be included. Penalized log likelihood maximization is used to fit the model together with the automatic smoothness selection.
scam为广义相加模型的组件功能的线性预测的GAM形状的限制下提供的功能。模型可以包含多个形状的约束和无约束条件,以及双或单单调二元平滑。模型组是在的不受约束的mgcv(gam)与添加的约束的形状平滑,所以的平滑可以是一个以上的变量,和其他用户定义平滑可以包含相同。判罚对数似然最大化来拟合模型,的自动平滑度选择。
A function extrapolate.uni.scam to predict future values of the response variable in case of a single univariate shape constrained term has been added. Also univariate smooths subject to convexity/concavity constraints are available now as model terms.
的功能extrapolate.uni.scam预测未来的值的响应变量的情况下,一个单一的单变量形状的约束项已被添加。此外,单变量平滑凸/凹约束模型计算。
Details
详细信息----------Details----------
The package provides generalized additive modelling under shape constraints on the component functions of the linear predictor. scam and plot.scam functions are based on the functions of the unconstrained GAM mgcv(gam) and mgcv(plot.gam) and similar in use. summary.scam allows to extract the results of the model fitting in the same way as in summary.gam. A Bayesian approach is used to obtain a covariance matrix of the model coefficients and credible intervals for each smooth.
该软件包提供的组件功能的线性预测的形状约束下的广义相加模型。 scam和plot.scam函数上不受约束的GAM mgcv(gam)和mgcv(plot.gam)和使用中的功能。 summary.scam允许在summary.gam中的相同的方式来提取结果的模型拟合。使用了贝叶斯方法获得的模型系数的协方差矩阵和可信区间为每光滑。
(作者)----------Author(s)----------
Natalya Pya <nat.pya@gmail.com> based partly on <code>mgcv</code> by Simon Wood
Maintainer: Natalya Pya <nat.pya@gmail.com>
参考文献----------References----------
additive models. J.R.Statist.Soc.B 70(3):495-518
linear models. Journal of the Royal Statistical Society: Series B. 73(1): 1–34