Miscellaneous Code and Programs
Dantzig estimation, imputation, and censored outcomes
Johnson BA, Long Q, and Conrad M. (2011) On path restoration for censored outcomes. Biometrics, 67, 1379-1388.
R code: bjds.tar.gz
Rank-based boosting in the accelerated failure time model
Johnson BA and Long Q. (2011) Survival ensembles by the sum of pairwise differences. Annals of Applied Statistics, In press.
R code: gehan_boost.tar.gz
Variable selection in the partly linear model for censored data
Johnson BA. (2009) Estimation in the L1-regularized partly linear model for censored outcomes with application to integrated analyses of clinical predictors and gene expression data. Biostatistics, 10, 640-658.
R code: plm_rank_l1.tar.gz
Rank-based variable selection for censored data, i.e., Gehan lasso
Johnson BA. (2009) Estimation in the L1-regularized partly linear model for censored outcomes with application to integrated analyses of clinical predictors and gene expression data. Biostatistics, 10, 640-658. Johnson BA. (2008) Estimation in the L1-regularized accelerated failure time model. Technical Report 08-01, Department of Biostatistics, Emory University, May 2008.
R code: gehan_lasso.tar.gz
Collection of R programs for heuristic dimension reduction by Cox score, then Gehan lasso
(MORENO microarray project):
R code: cmprogs.tar.gz
One-step Penalized Buckley-James statistics
Johnson BA. (2009) On lasso for censored data. Electronic Journal of Statistics, 3, 485-506.
R code: penbj.tar.gz
Rank-based variable selection
Johnson BA and Peng L. (2008) Rank-based variable selection. Journal of Nonparametric Statistics, 20, 241-252.
R code: rankvs.tar.gz
Johnson BA, Lin DY, and Zeng D. (2008) Penalized estimating functions and variable selection in semiparametric regression models. Journal of the American Statistical Association, 103, 672-680.
R code: coarse.tar.gz (All functions were written entirely in C and use the GSL library.)
Regression with informatively censored predictors
Johnson BA and Tsiatis AA. (2004) Estimating mean response as a function of treatment duration in an observational study, where duration may be informatively censored. Biometrics, 60, 315-323.
R code: ipw_mean.tar.gz