README.md

In WinVector/QSurvival: Quick survival models by data transformation

QSurvival

Quasi observation based survival package for R.

This package supplies an explicit quasi observation implementation of survival models.

Obviously there are already a lot of R packages that compute surival models (see here). This package is intended to help teach.

One can use this to reproduce Cox's discrete time logistic hazard model, or build a model with constant base hazard rate, or directly control how time enters the model. You can also change what machine learning algorithm is used make the hazard predictions. This is a teaching package showing some of the issues in survival modeling. The best way to approach this package is through the included vignettes (including showing the use of the mgcv package for survival modeling).

To install:

# install.packages('devtools')
devtools::install_github('WinVector/QSurvival', build_vignettes=TRUE)

WinVector/QSurvival documentation built on May 9, 2019, 10:59 p.m.