Learn R Programming

HQM (version 2.0)

Superefficient Estimation of Future Conditional Hazards Based on Marker Information

Description

Provides univariate and indexed (multivariate) nonparametric smoothed kernel estimators for the future conditional hazard rate function when time-dependent covariates are present, a bandwidth selector for the estimator's implementation and pointwise and uniform confidence bands. Methods used in the package refer to Bagkavos, Isakson, Mammen, Nielsen and Proust-Lima (2025) .

Copy Link

Version

Install

install.packages('HQM')

Monthly Downloads

292

Version

2.0

License

GPL (>= 2)

Maintainer

Dimitrios Bagkavos

Last Published

January 8th, 2026

Functions in HQM (2.0)

R_K

Bandwidth selection score R
b_selection_index_optim

Cross validation index parameter selection
Sim.True.Hazard

Simulated true hazard (bootstrap average)
dataset_split

Split dataset for K-fold cross validation
b_selection

Cross validation bandwidth selection
bs.hqm

Brier score for the High Quality Marker estimator
b_selection_prep_g

Preparations for bandwidth selection
auc.hqm

AUC for the High Quality Marker estimator
StudentizedBwB.Index.CIs

Compute Studentized Double Bootstrap Pointwise Confidence Intervals for the Indexed Hazard Rate Estimate
SingleIndCondFutHaz

Local linear future conditional hazard estimator (wrapper)
index_optim

Indexing parameter objective function
dij

D matrix entries, used for the implementation of the local linear kernel
lin_interpolate

Linear interpolation
h_xt

Local constant future conditional hazard rate estimation at a single time point
h_xtll

Local linear future conditional hazard rate estimation at a single time point
get_h_xll

Local linear future conditional hazard rate estimator
g_xt

Computation of a key component for wild bootstrap
h_xt_vec

Hqm estimator on the marker grid
get_h_x

Local constant future conditional hazard rate estimator
get_alpha

Marker-only hazard rate
llK_b

Local linear kernel
pbc2

Mayo Clinic Primary Biliary Cirrhosis Data
make_sf

Survival function from a hazard
llweights

Local linear weight functions
to_id

Event data frame
prep_cv2

Prepare for Cross validation index parameter selection
prep_boot

Precomputation for wild bootstrap
prep_cv

Prepare for Cross validation bandwidth selection
Q1

Bandwidth selection score Q1
Conf_bands

Confidence bands
Quantile.Index.CIs

Compute Quantile Pointwise Confidence Intervals for for the Indexed Hazard Rate Estimate
make_N, make_Ni, make_Y, make_Yi

Occurance and Exposure on grids
BwB.HRandIndex.param

Bootstrap Estimation of Hazard Function and Index Parameters
Kernels

Classical (unmodified) kernel and related functionals
Pivot.Index.CIs

Compute Pivot Pointwise Confidence Intervals for the Indexed Hazard Rate Estimate
Epan

Epanechnikov kernel
Boot.hqm

Indexed HQM hazard estimator for one bootstrap sample
Boot.hrandindex.param

Bootstrap Estimation of Hazard Function and Index Parameters