Learn R Programming

Introduction

fastcmprsk is an R package for performing Fine-Gray regression via a forward-backward scan algorithm.

Official release is available on CRAN and the master branch on GitHub.

Features

  • Scalable Fine-Gray estimation procedure for large-scale competing risks data.
  • Currently supports unpenalized and penalized (LASSO, ridge, SCAD, MCP, elastic-net) regression.
  • Can perform CIF estimation with interval/band estimation via bootstrap.

What’s New in Version 1.1.0?

  • Official version is loaded onto CRAN.

Implementation

fastcmprsk in an R package with most functionality implemented in C. The package uses cyclic coordinate descent to optimize the likelihood function.

Installation

To install the latest development version, install from GitHub.

install.packages("devtools")
devtools::install_github(“erickawaguchi/fastcmprsk”)

System Requirements

Requires R (version 3.5.0 or higher).

User Documentation

License

fastcmprsk is licensed under GPL-3.

Development

fastcmprsk is being developed in R Studio.

Copy Link

Version

Install

install.packages('fastcmprsk')

Monthly Downloads

665

Version

1.1.1

License

GPL-3

Maintainer

Eric Kawaguchi

Last Published

September 11th, 2019

Functions in fastcmprsk (1.1.1)

Crisk

Create a Competing Risk Object
varianceControl

Controls for Variance Calculation
coef.fcrrp

Extract coefficients from an "fcrrp" object.
coef.fcrr

Extract coefficients from an "fcrr" object.
summary.fcrr

Summary method for fastCrr
plot.predict.fcrr

Plots predicted cumulative incidence function
confint.fcrr

Confidence Intervals for Model Parameters
plot.fcrrp

Plots solution path for penalized methods
fastCrr

Fast Fine-Gray Model Estimation
vcov.fcrr

Extract variance-covariance matrix from an "fcrr" object.
fastCrrp

Penalized Fine-Gray Model Estimation via two-way linear scan
simulateTwoCauseFineGrayModel

Simulate data from the Fine-Gray Model
AIC.fcrr

Akaike's An Information Criterion
predict.fcrr

Cumulative Incidence Function Estimation
AIC.fcrrp

Akaike's An Information Criterion
print.summary.fcrr

Prints summary of a fcrr x
logLik.fcrr

Extract log-pseudo likelihood from an "fcrr" object.
logLik.fcrrp

Extract log-pseudo likelihood from an "fcrrp" object.