plsRcox (version 0.2.0)

plsRcox-package: Partial least squares Regression for Cox models and related techniques

Description

This packages provides Partial least squares Regression and various techniques for fitting Cox models in high dimensionnal settings. It allows for Kfold crossvalidation of such models using various criteria, missing data in the eXplanatory variables. Bootstrap confidence intervals constructions are also available.

Details

ll{ Package: plsRcox Version: 0.2.0 Date: 2011-01-13 Depends: R (>= 2.4.0) Imports: boot, plsRglm, lars, survival, pls Enhances: Suggests: glcoxph, survivalROC, plsRglm, lars, survival, pls License: GPL-3 Encoding: latin1 URL: http://www-irma.u-strasbg.fr/~fbertran/ Classification/MSC: 62N01, 62N02, 62N03, 62N99 Built: R 2.12.1; ; 2011-01-13 12:28:22 UTC; windows } Index: DR_coxph (Deviance) Residuals Computation Xmicro.censure_compl_imp Imputed Microsat features coxDKplsDR Fitting a Direct Kernel PLS model on the (Deviance) Residuals cox_glcoxph_supp_vals_KM Demo dataset cox_glcoxph_supp_vals_KM_micro Demo dataset cox_glcoxph_supp_vals_NNE Demo dataset cox_glcoxph_supp_vals_NNE_micro Demo dataset cox_pls2_supp_vals_KM Demo dataset cox_pls2_supp_vals_KM_micro Demo dataset cox_pls2_supp_vals_NNE Demo dataset cox_pls2_supp_vals_NNE_micro Demo dataset cox_plsDR2_supp_vals_KM Demo dataset cox_plsDR2_supp_vals_KM_micro Demo dataset cox_plsDR2_supp_vals_NNE Demo dataset cox_plsDR2_supp_vals_NNE_micro Demo dataset coxpls Fitting a Cox-Model on PLSR components coxpls2 Fitting a Cox-Model on PLSR components coxplsDR Fitting a PLSR model on the (Deviance) Residuals coxplsDR2 Fitting a PLSR model on the (Deviance) Residuals larsDR_coxph Fitting a LASSO/LARS model on the (Deviance) Residuals lars_supp_vals_KM Demo dataset lars_supp_vals_KM_micro Demo dataset lars_supp_vals_NNE Demo dataset lars_supp_vals_NNE_micro Demo dataset micro.censure Microsat features and survival times

References

plsRcox : mod�les{mod`eles} de Cox en pr�sence{pr'esence} d'un grand nombre de variables explicatives, Fr�d�ric{Fr'ed'eric} Bertrand, Myriam Maumy-Bertrand, Marie-Pierre Gaub, Nicolas Meyer, Chimiom�trie{Chimiom'etrie} 2010, Paris, 2010.

See Also

plsRglm

Examples

data(micro.censure)
Y_train_micro <- micro.censure$survyear[1:80]
C_train_micro <- micro.censure$DC[1:80]

DR_coxph(Y_train_micro,C_train_micro,plot=TRUE)
DR_coxph(Y_train_micro,C_train_micro,scaleY=FALSE,plot=TRUE)
DR_coxph(Y_train_micro,C_train_micro,scaleY=TRUE,plot=TRUE)

rm(Y_train_micro,C_train_micro)


data(micro.censure)
data(Xmicro.censure_compl_imp)

X_train_micro <- Xmicro.censure_compl_imp[1:80,]
Y_train_micro <- micro.censure$survyear[1:80]
C_train_micro <- micro.censure$DC[1:80]

(cox_larsDR_fit <- larsDR_coxph(X_train_micro,Y_train_micro,C_train_micro,max.steps=6,use.Gram=FALSE,scaleX=TRUE))
(cox_larsDR_fit <- larsDR_coxph(~X_train_micro,Y_train_micro,C_train_micro,max.steps=6,use.Gram=FALSE,scaleX=TRUE))
(cox_larsDR_fit <- larsDR_coxph(~.,Y_train_micro,C_train_micro,max.steps=6,use.Gram=FALSE,scaleX=TRUE,dataXplan=data.frame(X_train_micro)))

larsDR_coxph(~X_train_micro,Y_train_micro,C_train_micro,max.steps=6,use.Gram=FALSE)
larsDR_coxph(~X_train_micro,Y_train_micro,C_train_micro,max.steps=6,use.Gram=FALSE,scaleX=FALSE)
larsDR_coxph(~X_train_micro,Y_train_micro,C_train_micro,max.steps=6,use.Gram=FALSE,scaleX=TRUE,allres=TRUE)

rm(X_train_micro,Y_train_micro,C_train_micro,cox_larsDR_fit)


data(micro.censure)
data(Xmicro.censure_compl_imp)

X_train_micro <- Xmicro.censure_compl_imp[1:80,]
Y_train_micro <- micro.censure$survyear[1:80]
C_train_micro <- micro.censure$DC[1:80]

(cox_pls_fit <- coxpls(X_train_micro,Y_train_micro,C_train_micro,nt=7,typeVC="none"))
(cox_pls_fit <- coxpls(~X_train_micro,Y_train_micro,C_train_micro,nt=7,typeVC="none"))
(cox_pls_fit <- coxpls(~.,Y_train_micro,C_train_micro,nt=7,typeVC="none",data=data.frame(X_train_micro)))

rm(X_train_micro,Y_train_micro,C_train_micro,cox_pls_fit)


data(micro.censure)
data(Xmicro.censure_compl_imp)

X_train_micro <- Xmicro.censure_compl_imp[1:80,]
Y_train_micro <- micro.censure$survyear[1:80]
C_train_micro <- micro.censure$DC[1:80]

(cox_pls2_fit=coxpls2(X_train_micro,Y_train_micro,C_train_micro,ncomp=6,validation="CV"))
(cox_pls2_fit=coxpls2(~X_train_micro,Y_train_micro,C_train_micro,ncomp=6,validation="CV"))
(cox_pls2_fit=coxpls2(~.,Y_train_micro,C_train_micro,ncomp=6,validation="CV",dataXplan=data.frame(X_train_micro)))

rm(X_train_micro,Y_train_micro,C_train_micro,cox_pls2_fit)


data(micro.censure)
data(Xmicro.censure_compl_imp)

X_train_micro <- Xmicro.censure_compl_imp[1:80,]
Y_train_micro <- micro.censure$survyear[1:80]
C_train_micro <- micro.censure$DC[1:80]

(cox_plsDR_fit <- coxplsDR(X_train_micro,Y_train_micro,C_train_micro,nt=7))
(cox_plsDR_fit <- coxplsDR(~X_train_micro,Y_train_micro,C_train_micro,nt=7))
(cox_plsDR_fit <- coxplsDR(~.,Y_train_micro,C_train_micro,nt=7,dataXplan=data.frame(X_train_micro)))

rm(X_train_micro,Y_train_micro,C_train_micro,cox_plsDR_fit)


data(micro.censure)
data(Xmicro.censure_compl_imp)

X_train_micro <- Xmicro.censure_compl_imp[1:80,]
Y_train_micro <- micro.censure$survyear[1:80]
C_train_micro <- micro.censure$DC[1:80]

(cox_plsDR2_fit=coxplsDR2(X_train_micro,Y_train_micro,C_train_micro,ncomp=6,validation="none"))
(cox_plsDR2_fit=coxplsDR2(~X_train_micro,Y_train_micro,C_train_micro,ncomp=6,validation="none"))
(cox_plsDR2_fit=coxplsDR2(~.,Y_train_micro,C_train_micro,ncomp=6,validation="none",dataXplan=data.frame(X_train_micro)))

rm(X_train_micro,Y_train_micro,C_train_micro,cox_plsDR2_fit)


data(micro.censure)
data(Xmicro.censure_compl_imp)

X_train_micro <- Xmicro.censure_compl_imp[1:80,]
Y_train_micro <- micro.censure$survyear[1:80]
C_train_micro <- micro.censure$DC[1:80]

(cox_DKplsDR_fit=coxDKplsDR(X_train_micro,Y_train_micro,C_train_micro,ncomp=6,validation="CV"))
(cox_DKplsDR_fit=coxDKplsDR(~X_train_micro,Y_train_micro,C_train_micro,ncomp=6,validation="CV"))
(cox_DKplsDR_fit=coxDKplsDR(~.,Y_train_micro,C_train_micro,ncomp=6,validation="CV",dataXplan=data.frame(X_train_micro)))

(cox_DKplsDR_fit=coxDKplsDR(X_train_micro,Y_train_micro,C_train_micro,ncomp=6,validation="CV",allres=TRUE))
(cox_DKplsDR_fit=coxDKplsDR(~X_train_micro,Y_train_micro,C_train_micro,ncomp=6,validation="CV",allres=TRUE))
(cox_DKplsDR_fit=coxDKplsDR(~.,Y_train_micro,C_train_micro,ncomp=6,validation="CV",allres=TRUE,dataXplan=data.frame(X_train_micro)))

rm(X_train_micro,Y_train_micro,C_train_micro,cox_DKplsDR_fit)