dynamichazard: Dynamic Hazard Models using State Space Models

Contains functions that lets you fit dynamic hazard models with binary outcomes using state space models. The methods are originally described in Fahrmeir (1992) <doi:10.1080/01621459.1992.10475232> and Fahrmeir (1994) <doi:10.1093/biomet/81.2.317>. The functions also provide an extension hereof where the Extended Kalman filter is replaced by an Unscented Kalman filter. Models are fitted with the regular coxph() like formula.

Version: 0.2.0
Depends: R (≥ 3.3.2), stats, graphics, utils, survival
Imports: parallel, Rcpp (≥ 0.12.6), boot, stringr
LinkingTo: Rcpp, RcppArmadillo
Suggests: testthat, knitr, rmarkdown, timereg, captioner, biglm, httr, mgcv, shiny, formatR, R.rsp
Published: 2017-01-20
Author: Benjamin Christoffersen [cre, aut], Alan Miller [ctb], Anthony Williams [ctb], Boost developers [ctb, cph], R-core [ctb, cph]
Maintainer: Benjamin Christoffersen <boennecd at gmail.com>
BugReports: https://github.com/boennecd/dynamichazard/issues
License: GPL-2
URL: https://github.com/boennecd/dynamichazard
NeedsCompilation: yes
SystemRequirements: C++11
Materials: NEWS
CRAN checks: dynamichazard results


Reference manual: dynamichazard.pdf
Vignettes: Comparing methods for time varying logistic models
Bootstrap illustration
Simulation study with logit model
Package source: dynamichazard_0.2.0.tar.gz
Windows binaries: r-devel: dynamichazard_0.2.0.zip, r-release: dynamichazard_0.2.0.zip, r-oldrel: not available
OS X Mavericks binaries: r-release: dynamichazard_0.2.0.tgz, r-oldrel: not available
Old sources: dynamichazard archive


Please use the canonical form https://CRAN.R-project.org/package=dynamichazard to link to this page.