CP_SIR: Counting process based sliced inverse regression model

In orthoDr: Semi-Parametric Dimension Reduction Models Using Orthogonality Constrained Optimization

Counting process based sliced inverse regression model

Description

The CP-SIR model for right-censored survival outcome. This model is correct only under very strong assumptions, however, since it only requires an SVD, the solution is used as the initial value in the orthoDr optimization.

Usage

CP_SIR(x, y, censor, bw = silverman(1, length(y)))

Arguments

x	A matrix for features (continuous only).
y	A vector of observed time.
censor	A vector of censoring indicator.
bw	Kernel bandwidth for nonparametric estimations (one-dimensional), the default is using Silverman's formula.

Value

A list consisting of

values	The eigenvalues of the estimation matrix
vectors	The estimated directions, ordered by eigenvalues

References

Sun, Q., Zhu, R., Wang, T. and Zeng, D. (2019) "Counting Process Based Dimension Reduction Method for Censored Outcomes." Biometrika, 106(1), 181-196. DOI: \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1093/biomet/asy064")}

Examples

# This is setting 1 in Sun et. al. (2017) with reduced sample size
library(MASS)
set.seed(1)
N <- 200
P <- 6
V <- 0.5^abs(outer(1:P, 1:P, "-"))
dataX <- as.matrix(mvrnorm(N, mu = rep(0, P), Sigma = V))
failEDR <- as.matrix(c(1, 0.5, 0, 0, 0, rep(0, P - 5)))
censorEDR <- as.matrix(c(0, 0, 0, 1, 1, rep(0, P - 5)))
T <- rexp(N, exp(dataX %*% failEDR))
C <- rexp(N, exp(dataX %*% censorEDR - 1))
ndr <- 1
Y <- pmin(T, C)
Censor <- (T < C)

# fit the model
cpsir.fit <- CP_SIR(dataX, Y, Censor)
distance(failEDR, cpsir.fit$vectors[, 1:ndr, drop = FALSE], "dist")

orthoDr documentation built on April 30, 2023, 5:12 p.m.