Skip to content

Latest commit

 

History

History
65 lines (49 loc) · 2.15 KB

README.md

File metadata and controls

65 lines (49 loc) · 2.15 KB
Raw

rankIC: R package for fitting partially interval-censored rank regressions

Overview

rankIC is the R package to fit the linear rank regressions when the data are partially interval-censored.

Installation

devtools::install_github(repo='taehwa015/rankIC')

Status of development

The code provided here is only being provided for research purposes.

Documentation

Vignette is available at here.

Usage

The rankIC package provides a rank-based estimating procedures for the partially interval-censored data. See Choi et al. (2022+) for more detailed description of the method.

Below example is the phase 3 metastatic colorectal cancer clinical trial study. Since the event times are possibly correlated within same patient, we regard patient id as the cluster. To adjust informative cluster sizes, we further consider weight function, motivated by Wang and Zhao (2008). Larger cluster will be underweighted by letting alpha = 1, while cluster structure will be ignored alpha = 0. Note that by letting id = NULL, we fit the univariate AFT model.

library(PICBayes)
data(mCRC)
dt0 = as.data.frame(mCRC)
d = with(dt0,
         data.frame(U = ifelse(is.na(L), 0, L),
                    V = ifelse(is.na(R), Inf, R),
                    Delta = 1-IC,
                    x1 = TRT_C,
                    x2 = KRAS_C,
                    id = SITE))
U = d$U; V = d$V; X = cbind(d$x1, d$x2); Delta = d$Delta; id = d$id
aft_rank(U = U, V = V, X = X, Delta = Delta, id = id, 
         alpha = 1, type = "gehan", R = 10)
aft_rank(U = U, V = V, X = X, Delta = Delta, id = id, 
         alpha = 1, type = "logrank", R = 10)

Reference

  • Choi, T., Choi, S. and Bandyopadhyay, D. (2022+). Rank estimation for the accelerated failure time model with partially interval-censored data, (Under Review)

  • Wang, Y. G. and Zhao, Y. (2008). Weighted rank regression for clustered data analysis. Biometrics, 64(1), 39--45.

  • Pan, C. (2021). PICBayes: Bayesian Models for Partly Interval-Censored Data. R package. https://CRAN.R-project.org/package=PICBayes.