fit.lspls.glm: Fitting LS-PLS for generalized model for logistic regression
In lsplsGlm: Classification using LS-PLS for Logistic Regression

Description Usage Arguments Details Value Author(s) References See Also Examples

Fits a Least Square-Partial Least Square for logistic regression model. There are 3 extensions.

1
2
3

fit.lspls.glm(Y, X, D, W=diag(rep(1,nrow(D))), ncomp, method=c("LS-PLS-IRLS", 
"R-LS-PLS", "IR-LS-PLS"),lambda=NULL, penalized = NULL, nbrIterMax = NULL, 
threshold= NULL)

`Y`	a vector of length `n` giving the classes of the `n` observations. The classes must be coded as 1 or 0.
`X`	a data matrix (`nxp`) of genes. NAs and Inf are not allowed. Each row corresponds to an observation and each column to a gene.
`D`	a data matrix (`nxq`) of clinical data. NAs and Inf are not allowed. Each row corresponds to an observation and each column to a clinical variable.
`W`	weight matrix. If `W` is the identity matrix then the function will fit a standard LS-PLS model.
`ncomp`	a positive integer. `ncomp` is the number of selected components.
`method`	one of the 3 extensions of LS-PLS for logistic regression models.
`penalized`	if TRUE the parameter associated with `D` is ridge penalized. To use only if `method` equals to "R-LS-PLS".
`lambda`	coefficient of ridge penalty. If `penalized = TRUE`, `lambda` is the penalty for `D`. To use only if `method` equals to "R-LS-PLS".
`nbrIterMax`	maximal number of iterations. To use only if `method` equals to "R-LS-PLS" or "IR-LS-PLS".
`threshold`	used for the stopping rule. To use only if `method` equals to "R-LS-PLS" or "IR-LS-PLS".

This function fits LS-PLS models. With the argument "method" the user can choose one of the three extensions of LS-PLS for logistic regression (LS-PLS-IRLS, R-LS-PLS, IR-LS-PLS). For more details see references.

`coefficients`	vector of length `q+p` associate to `cbind(D,X)`.
`cvg`	the 0-1 value indicating convergence of the algorithm (1 for convergence, 0 otherwise).
`orthCoef`	coefficients matrix (`pxq`) returned also in the function `lspls` to be used to compute new predictors.
`projection`	the projection matrix used to convert `X` to scores.
`intercept`	the constant of the model.

Caroline Bazzoli, Thomas Bouleau, Sophie Lambert-Lacroix

Caroline Bazzoli, Sophie Lambert-Lacroix. Classification using LS-PLS with logistic regression based on both clinical and gene expression variables. 2017. <hal-01405101>

cv.lspls.glm,predict.lspls.glm.

#Data
data(BreastCancer)
#vector of responses
Y<-BreastCancer$Y
#Genetic data
X<-BreastCancer$X
#Clinical data
D<-BreastCancer$D
#SIS selection
X<-scale(X)

X<-SIS.selection(X=X,Y=Y,pred=50)
#Cross validation, 90% of our datasets is used to compose learning samples

#method LS-PLS-IRLS
ncompopt.lsplsirls<-cv.lspls.glm(Y=Y,X=X,D=D,folds=5,ncompmax=5,proportion=0.9,
		                                 method="LS-PLS-IRLS")$ncompopt
#method R-LS-PLS
cv<-cv.lspls.glm(Y=Y,X=X,D=D,ncompmax=5,proportion=0.9,method="R-LS-PLS",
	           lambda.grid=exp(log(10^seq(-3,2,0.7))),penalized=TRUE,
	           nbrIterMax=15,threshold=10^(-12))
ncompopt.rlspls<-cv$ncompopt
lambdaopt.rlspls<-cv$lambdaopt
#method IR-LS-PLS
ncompopt.irlspls<-cv.lspls.glm(Y=Y,X=X,D=D,ncompmax=5,proportion=0.9,
	method="IR-LS-PLS",nbrIterMax=15,threshold=10^(-12))$ncompopt         

#fitting model
fit.lsplsirls<-fit.lspls.glm(Y=Y,X=X,D=D,ncomp=ncompopt.lsplsirls,method="LS-PLS-IRLS")
fit.rlspls<-fit.lspls.glm(Y=Y,X=X,D=D,ncomp=ncompopt.rlspls,method="R-LS-PLS",
		         lambda=lambdaopt.rlspls,penalized=TRUE,nbrIterMax=15,
                         threshold=10^(-12))
fit.irlspls<-fit.lspls.glm(Y=Y,X=X,D=D,ncomp=ncompopt.irlspls,method="IR-LS-PLS",
                                                nbrIterMax=15,threshold=10^(-12))