A generic class for regression objects with network cohesion.
Objects can be created by calls of the form
The individual effects of the regression.
The fixed effects or covariate coefficients of the regression.
The network adjacency matrix for which cohession is assumed.
Parameter for cohesion penalty.
The response data frame with the first column being the observed time and the second column being the event indicator.
Regularization parameter for graph Laplacian.
Number of folds in cross-validation.
Cross-validated prediciton loss. It is MSE for linear regression, binomial deviance for logistic regression and test partial loglikelihood for Cox's model (see reference paper).
Standard deviation of cross-validation loss. It can be used for cross-validation by 1 sigma rule. It is more robust to noises.
The specific regression model used.
No methods defined with class "rncReg" in the signature.
Tianxi Li, Elizaveta Levina, Ji Zhu
Maintainer: Tianxi Li email@example.com
Tianxi Li, Elizaveta Levina and Ji Zhu. (2016)
Regression with network cohesion,
Verweij, Pierre JM, and Hans C. Van Houwelingen. (1993) Cross-validation in survival analysis, Statistics in medicine 12, no. 24: 2305-2314.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.