pseval: Methods for Evaluating Principal Surrogates of Treatment Response

Contains the core methods for the evaluation of principal surrogates in a single clinical trial. Provides a flexible interface for defining models for the risk given treatment and the surrogate, the models for integration over the missing counterfactual surrogate responses, and the estimation methods. Estimated maximum likelihood and pseudo-score can be used for estimation, and the bootstrap for inference. A variety of post-estimation summary methods are provided, including print, summary, plot, and testing.

Install the latest version of this package by entering the following in R:
AuthorMichael C Sachs [aut, cre], Erin E Gabriel [aut]
Date of publication2016-09-23 22:53:05
MaintainerMichael C Sachs <>
LicenseMIT + file LICENSE

View on CRAN

Man pages

add_bootstrap: Bootstrap resampling parameters

add_estimate: Estimate parameters

add_integration: Integration models

add_riskmodel: Add risk model to a psdesign object

calc_risk: Calculate the risk and functions of the risk

calc_STG: Calculate the Standardized total gain

empirical_TE: Compute the empirical Treatment Efficacy

empirical_VE: Compute the empirical Treatment Efficacy

expand_augdata: Expand augmented data using the integration function

generate_example_data: Generate sample data used for testing

integrate_bivnorm: Bivariate normal integration models for the missing S(1)

integrate_nonparametric: Nonparametric integration model for the missing S(1)

integrate_parametric: Parametric integration model for the missing S(1)

integrate_semiparametric: Semiparametric integration model using the location-scale...

plot.psdesign: Plot summary statistics for a psdesign object Modify a psdesign object by adding on new components.

print.psdesign: Concisely print information about a psdesign object

ps_bootstrap: Estimate parameters from a specified model using bootstrap...

psdesign: Specify a design for a principal surrogate evaluation

ps_estimate: Estimate parameters from a specified model using estimated...

pseudo_score: Estimate parameters from a specified model using pseudo-score

risk_binary: Risk model for binary outcome

riskcalc: Calculate risks with handlers for survival data

risk_continuous: Risk model for continuous outcome

risk_exponential: Exponential risk model for time to event outcome

risk.logit: Logit link function

risk_poisson: Poisson risk model for count outcomes

risk.probit: Probit link function

risk_weibull: Weibull risk model for time to event outcome

sp_locscale: Fit the semi-parametric location-scale model

stg: Compute the standardized total gain

summarize_bs: Summarize bootstrap samples

summary.psdesign: Summary method for psdesign objects

TE: Treatment efficacy contrast functions

verify_trt: Check that a variable is suitable for using as binary...

wem_test: Test for wide effect modificiation


add_bootstrap Man page
add_estimate Man page
add_integration Man page
add_riskmodel Man page
calc_risk Man page
calc_STG Man page
empirical_TE Man page
empirical_VE Man page
expand_augdata Man page
generate_example_data Man page
integrate_bivnorm Man page
integrate_nonparametric Man page
integrate_parametric Man page
integrate_semiparametric Man page
plot.psdesign Man page
print.psdesign Man page Man page
ps_bootstrap Man page
psdesign Man page
ps_estimate Man page
pseudo_score Man page
risk_binary Man page
riskcalc Man page
risk_continuous Man page
risk_exponential Man page
risk.logit Man page
risk_poisson Man page
risk.probit Man page
risk_weibull Man page
sp_locscale Man page
stg Man page
summarize_bs Man page
summary.psdesign Man page
TE Man page
verify_trt Man page
wem_test Man page

Questions? Problems? Suggestions? or email at

Please suggest features or report bugs with the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.