R/RcppExports.R

Defines functions Hess_rho grad_rho deriv_R colSums2

Documented in colSums2 deriv_R grad_rho Hess_rho

# Generated by using Rcpp::compileAttributes() -> do not edit by hand
# Generator token: 10BE3573-1514-4C36-9D1C-5A225CD40393

#' Matrix multiplication between two matrices
#'
#' @param Mat1 a matrix.
#' @param Mat2 another matrix.
#' @return prod the product Mat1%*%Mat2
#' @export
'%mult%' <- function(Mat1, Mat2) {
    .Call('_survPen_multmat', PACKAGE = 'survPen', Mat1, Mat2)
}

#' Matrix cross-multiplication between two matrices
#'
#' @param Mat1 a matrix.
#' @param Mat2 another matrix.
#' @return prod the product t(Mat1)%*%Mat2
#' @export
'%cross%' <- function(Mat1, Mat2) {
    .Call('_survPen_multcross', PACKAGE = 'survPen', Mat1, Mat2)
}

#' Matrix multiplication between a matrix and a vector
#'
#' @param Mat a matrix.
#' @param vec a vector.
#' @return prod the product Mat%*%vec
#' @export
'%vec%' <- function(Mat, vec) {
    .Call('_survPen_multvec', PACKAGE = 'survPen', Mat, vec)
}

#' colSums of a matrix
#'
#' @param Mat a matrix.
#' @return colSums(Mat)
#' @export
colSums2 <- function(Mat) {
    .Call('_survPen_colSums2', PACKAGE = 'survPen', Mat)
}

#' Derivative of a Choleski factor
#'
#' @param deriv_Vp derivatives of the Bayesian covariance matrix wrt rho (log smoothing parameters).
#' @param p number of regression parameters
#' @param R1 Choleski factor of Vp
#' @return a list containing the derivatives of R1 wrt rho (log smoothing parameters)
#' @export
deriv_R <- function(deriv_Vp, p, R1) {
    .Call('_survPen_deriv_R', PACKAGE = 'survPen', deriv_Vp, p, R1)
}

#' Gradient vector of LCV and LAML wrt rho (log smoothing parameters)
#'
#' @param X_GL list of matrices (\code{length(X.GL)=n.legendre}) for Gauss-Legendre quadrature
#' @param GL_temp list of vectors used to make intermediate calculations and save computation time
#' @param haz_GL list of all the matrix-vector multiplications X.GL[[i]]\%*\%beta for Gauss Legendre integration in order to save computation time
#' @param deriv_rho_beta firt derivative of beta wrt rho (implicit differentiation)
#' @param weights vector of weights for Gauss-Legendre integration on [-1;1]
#' @param tm vector of midpoints times for Gauss-Legendre integration; tm = 0.5*(t1 - t0)
#' @param nb_smooth number of smoothing parameters
#' @param p number of regression parameters
#' @param n_legendre number of nodes for Gauss-Legendre quadrature
#' @param S_list List of all the rescaled penalty matrices multiplied by their associated smoothing parameters
#' @param temp_LAML temporary matrix used when method="LAML" to save computation time
#' @param Vp Bayesian covariance matrix
#' @param S_beta List such that S_beta[[i]]=S_list[[i]]\%*\%beta
#' @param beta vector of estimated regression parameters
#' @param inverse_new_S inverse of the penalty matrix
#' @param X design matrix for the model
#' @param temp_deriv3 temporary matrix for third derivatives calculation when type="net" to save computation time
#' @param event vector of right-censoring indicators
#' @param expected vector of expected hazard rates
#' @param type "net" or "overall"
#' @param Ve frequentist covariance matrix
#' @param mat_temp temporary matrix used when method="LCV" to save computation time
#' @param method criterion used to select the smoothing parameters. Should be "LAML" or "LCV"; default is "LAML"
#' @return List of objects with the following items:
#' \item{grad_rho}{gradient vector of LCV or LAML}
#' \item{deriv_rho_inv_Hess_beta}{List of first derivatives of Vp wrt rho}
#' \item{deriv_rho_Hess_unpen_beta}{List of first derivatives of the Hessian of the unpenalized log-likelihood wrt rho}
#' @export
grad_rho <- function(X_GL, GL_temp, haz_GL, deriv_rho_beta, weights, tm, nb_smooth, p, n_legendre, S_list, temp_LAML, Vp, S_beta, beta, inverse_new_S, X, temp_deriv3, event, expected, type, Ve, mat_temp, method) {
    .Call('_survPen_grad_rho', PACKAGE = 'survPen', X_GL, GL_temp, haz_GL, deriv_rho_beta, weights, tm, nb_smooth, p, n_legendre, S_list, temp_LAML, Vp, S_beta, beta, inverse_new_S, X, temp_deriv3, event, expected, type, Ve, mat_temp, method)
}

#' Hessian matrix of LCV and LAML wrt rho (log smoothing parameters)
#'
#' @param X_GL list of matrices (\code{length(X.GL)=n.legendre}) for Gauss-Legendre quadrature
#' @param X_GL_Q list of transformed matrices from X_GL in order to calculate only the diagonal of the fourth derivative of the likelihood
#' @param GL_temp list of vectors used to make intermediate calculations and save computation time
#' @param haz_GL list of all the matrix-vector multiplications X.GL[[i]]\%*\%beta for Gauss Legendre integration in order to save computation time
#' @param deriv2_rho_beta second derivatives of beta wrt rho (implicit differentiation)
#' @param deriv_rho_beta firt derivatives of beta wrt rho (implicit differentiation)
#' @param weights vector of weights for Gauss-Legendre integration on [-1;1]
#' @param tm vector of midpoints times for Gauss-Legendre integration; tm = 0.5*(t1 - t0)
#' @param nb_smooth number of smoothing parameters
#' @param p number of regression parameters
#' @param n_legendre number of nodes for Gauss-Legendre quadrature
#' @param deriv_rho_inv_Hess_beta list of first derivatives of Vp wrt rho
#' @param deriv_rho_Hess_unpen_beta list of first derivatives of Hessian of unpenalized log likelihood wrt rho
#' @param S_list List of all the rescaled penalty matrices multiplied by their associated smoothing parameters
#' @param minus_eigen_inv_Hess_beta vector of eigenvalues of Vp
#' @param temp_LAML temporary matrix used when method="LAML" to save computation time
#' @param temp_LAML2 temporary matrix used when method="LAML" to save computation time
#' @param Vp Bayesian covariance matrix
#' @param S_beta List such that S_beta[[i]]=S_list[[i]]\%*\%beta
#' @param beta vector of estimated regression parameters
#' @param inverse_new_S inverse of the penalty matrix
#' @param X design matrix for the model
#' @param X_Q transformed design matrix in order to calculate only the diagonal of the fourth derivative of the likelihood
#' @param temp_deriv3 temporary matrix for third derivatives calculation when type="net" to save computation time
#' @param temp_deriv4 temporary matrix for fourth derivatives calculation when type="net" to save computation time
#' @param event vector of right-censoring indicators
#' @param expected vector of expected hazard rates
#' @param type "net" or "overall"
#' @param Ve frequentist covariance matrix
#' @param deriv_rho_Ve list of derivatives of Ve wrt rho
#' @param mat_temp temporary matrix used when method="LCV" to save computation time
#' @param deriv_mat_temp list of derivatives of mat_temp wrt rho
#' @param eigen_mat_temp vector of eigenvalues of mat_temp
#' @param method criterion used to select the smoothing parameters. Should be "LAML" or "LCV"; default is "LAML"
#' @return Hessian matrix of LCV or LAML wrt rho
#' @export
Hess_rho <- function(X_GL, X_GL_Q, GL_temp, haz_GL, deriv2_rho_beta, deriv_rho_beta, weights, tm, nb_smooth, p, n_legendre, deriv_rho_inv_Hess_beta, deriv_rho_Hess_unpen_beta, S_list, minus_eigen_inv_Hess_beta, temp_LAML, temp_LAML2, Vp, S_beta, beta, inverse_new_S, X, X_Q, temp_deriv3, temp_deriv4, event, expected, type, Ve, deriv_rho_Ve, mat_temp, deriv_mat_temp, eigen_mat_temp, method) {
    .Call('_survPen_Hess_rho', PACKAGE = 'survPen', X_GL, X_GL_Q, GL_temp, haz_GL, deriv2_rho_beta, deriv_rho_beta, weights, tm, nb_smooth, p, n_legendre, deriv_rho_inv_Hess_beta, deriv_rho_Hess_unpen_beta, S_list, minus_eigen_inv_Hess_beta, temp_LAML, temp_LAML2, Vp, S_beta, beta, inverse_new_S, X, X_Q, temp_deriv3, temp_deriv4, event, expected, type, Ve, deriv_rho_Ve, mat_temp, deriv_mat_temp, eigen_mat_temp, method)
}

Try the survPen package in your browser

Any scripts or data that you put into this service are public.

survPen documentation built on Sept. 14, 2023, 1:06 a.m.