nlpred1.0.1 package

Estimators of Non-Linear Cross-Validated Risks Optimized for Small Samples

lpo_auc

Compute the leave-pair-out cross-validation estimator of AUC.

one_boot_auc

Internal function used to perform one bootstrap sample. The function `...

one_boot_scrnp

Internal function used to perform one bootstrap sample. The function `...

print.cvauc

Print results of cv_auc

print.scrnp

Print results of cv_scrnp

boot_auc

Compute the bootstrap-corrected estimator of AUC.

boot_scrnp

Compute the bootstrap-corrected estimator of SCRNP.

ci.cvAUC_withIC

ci.cvAUC_withIC

cv_auc

Estimates of CVAUC

cv_scrnp

Estimates of CV SCNP

dot-Dy

Compute one of the terms of the efficient influence function

dot-estim_fn

An estimating function for cvAUC

dot-estim_fn_nested_cv

An estimating function for cvAUC with initial estimates generated via ...

dot-get_auc

Compute the AUC given the cdf and pdf of psi

dot-get_cv_estim

Helper function to turn prediction_list into CV estimate of SCRNP

dot-get_density

Function to estimate density needed to evaluate standard errors.

dot-get_nested_cv_quantile

Helper function to get quantile for a single training fold data when n...

dot-get_one_fold

Helper function to get results for a single cross-validation fold

dot-get_predictions

Worker function for fitting prediction functions (possibly in parallel...

dot-get_psi_distribution

Compute the conditional (given Y = y) estimated distribution of psi

dot-get_psi_distribution_nested_cv

Compute the conditional (given Y = y) CV-estimated distribution of psi

dot-get_quantile

Helper function to get quantile for a single training fold data when n...

dot-make_long_data

Worker function to make long form data set needed for CVTMLE targeting...

dot-make_long_data_nested_cv

Worker function to make long form data set needed for CVTMLE targeting...

dot-make_targeting_data

Helper function for making data set in proper format for CVTMLE

dot-process_input

Unexported function from cvAUC package

F_nBn_star

Compute the targeted conditional cumulative distribution of the learne...

F_nBn_star_nested_cv

Compute the targeted conditional cumulative distribution of the learne...

fluc_mod_optim_0

Helper function for CVTMLE grid search

fluc_mod_optim_1

Helper function for CVTMLE grid search

glm_wrapper

Wrapper for fitting a logistic regression using glm.

glmnet_wrapper

Wrapper for fitting a lasso using package glmnet.

randomforest_wrapper

Wrapper for fitting a random forest using randomForest .

ranger_wrapper

Wrapper for fitting a random forest using ranger .

stepglm_wrapper

Wrapper for fitting a forward stepwise logistic regression using glm...

superlearner_wrapper

Wrapper for fitting a super learner based on SuperLearner.

xgboost_wrapper

Wrapper for fitting eXtreme gradient boosting via xgboost

Methods for obtaining improved estimates of non-linear cross-validated risks are obtained using targeted minimum loss-based estimation, estimating equations, and one-step estimation (Benkeser, Petersen, van der Laan (2019), <doi:10.1080/01621459.2019.1668794>). Cross-validated area under the receiver operating characteristics curve (LeDell, Petersen, van der Laan (2015), <doi:10.1214/15-EJS1035>) and other metrics are included.

  • Maintainer: David Benkeser
  • License: MIT + file LICENSE
  • Last published: 2020-02-23