Estimation functions

estimate_ipwrisk()

Estimates the cumulative risk with time-varying predictors of censoring using inverse-probability weighted estimating functions

estimate_gcomprisk()

Estimates the cumulative risk of a right-censored outcome using a g-computation estimator.

estimate_aipwrisk()

Estimates the cumulative risk of a right-censored outcome using an augmented inverse probability weighted estimator.

re_estimate()

A generic function to refit/re-estimate parameters described in a result object

compare_fits()

compare_fits: a function for computing the difference and 95% confidence interval between fit object point estimates using bootstrap estimates to generate a standard error

Table-making functions

make_table1()

Makes a unweighted or IP weighted Table 1

make_table2()

make_table2 is a generic function to create table 2's from cumrisk or cumcount objects

make_wt_summary_table()

Summarizes weights from a set of objects that inherit from class ipw

extreme_weights()

Creates a table of the most extreme weights from an ipw object

Plotting functions

plot(<cumrisk>)

Plots a panel of (possibly overlayed) cumulative risk functions or cumulative risk differences and confidence intervals.

hist(<ipw>)

Plots a panel of propensity score histograms

forest_plot(<cumrisk>)

Produces a forest plot of cumulative risks or measures of effect

Model specification functions

specify_models()

Specifies the data structure and models for a data set

identify_outcome()

Helper function for identifying an outcome variable for a model_specification object

identify_censoring()

Helper function for identifying a censoring variable for a model_specification object

identify_treatment()

Helper function for identifying a treatmement for a model_specification object

identify_competing_risk()

Helper function for adding a competing risk feature to a model_specification object

identify_subject()

Helper function for adding a subject identifier feature to a model_specification object

identify_interval()

Helper function for identifying start and stop times for observations in a longitudinal data structure

identify_missing()

Helper function for activating missing data weighting for a model_specification object

Functions to modify model specification in a result object

update_treatment()

Updates the treatment information in a result object for later re-estimation

update_censoring()

Updates the censoring information in a result object for later re-estimation

update_outcome()

Updates the outcome information in a result object for later re-estimation

update_missing()

Updates the missingness information in a result object for later re-estimation

subgroup()

Updates a result object with a subgroup for later re-estimation.

Miscellaneous functions used internally

StatStepribbon

ggplot2 geom to create stairstep confidence bands for cumulative risk estimates

boot_vector_create()

Function that creates a vector of bootstrap resampling weights

check_arguments()

Checks standard arguments to estimation functions

compare_protocols()

compare_protocols: a function for estimating cululative risk across a list of treatment protocols

compute_cumcount_ci()

Computes confidence intervals for cumulative count estimates

compute_cumcount_effect_ci()

Computes confidence intervals for functions of cumulative count estimates

compute_cumrisk_ci()

Computes confidence intervals for cumulative risk estimates

compute_cumrisk_effect_ci()

Computes confidence intervals for functions of cumulative risk estimates

create_censoring_indicators()

Builds a matrix of censoring indicators for a set of variables.

create_censoring_indicators_timevar_old()

Builds a matrix of censoring indicators for a set of variables.

cumriskdif_ci()

Computes cumulative risk differences and confidence intervals

cumriskrr_ci()

Computes cumulative risk differences and confidence intervals

effect_measure_select()

Returns an effect measure object given a supplied character string

estimate_aipwrisk_efficient()

Estimates the cumulative risk of a right-censored outcome using

estimate_censor_coxph()

estimate_censor_coxph: a function to estimate a censoring probability using coxph

estimate_censor_cph()

estimate_censor_cph: a function to estimate a censoring probability using cph

estimate_censoring_prob()

estimate_censoring_prob: a function to estimate a probability a patient will be uncensored

estimate_ipwcount()

Estimates the cumulative count with time-varying predictors of censoring using inverse-probability weighted estimating functions

estimate_ipweights()

Estimates the inverse_probability weights when no outcome is specified

estimate_missing_glmnet()

estimate_missing_glmnet: a function to estimate a missingness probability using logistic regression

estimate_missing_prob()

estimate_missing_prob: a function to estimate a missingness probability

estimate_outcome_coxph()

estimate_outcome_coxph: a function to estimate outcome probabilities at times in new data using coxph

estimate_outcome_cph()

estimate_outcome_cph: a function to estimate an outcome risk using cph function in R

estimate_ps()

estimate_ps: a function to estimate a propensity score

estimate_ps_glm()

estimate_ps_glm: a function to estimate a propensity score using glm logistic regression

estimate_ps_glmnet()

estimate_ps_glmnet: a function to estimate a propensity score using logistic regression

estimate_rates()

Estimates crude event rates

find_covs()

find_covs: a function to identify covariates used in a model

find_empty()

find_empty: a function to identify factors that only have individuals in a single level

find_missing()

find_missing: a function to identify rows with missing data and variables that have missing values

fit_coxph()

fit_coxph: a function to fit a coxph model

forest_plot()

A generic function to produce of forest plot estimates from results objects

identify_count()

Helper function for identifying a count variable for a model_specification object

initializer()

Standard set-up needed by estimating functions

is.aipw()

is.aipw: a function for checking whether an object inherits from aipw

is.boot_est()

is.boot_est: a function for checking whether an object inherits from boot_est

is.cumcount()

is.cumcount: a function for checking whether an object inherits from cumcount

is.cumrisk()

is.cumrisk: a function for checking whether an object inherits from cumrisk

is.gcomp()

is.gcomp: a function for checking whether an object inherits from gcomp

is.ipw()

is.ipw: a function for checking whether an object inherits from ipw

is.model_specification()

is.model_specification: a function for checking whether an object inherits from model_specification

logit_ps_est()

A function to estimate a propensity score using logistic regression

make_ipw_table1_sd()

Creates a standard weighted table 1 with standardized differences from a cumrisk object

named_list() named_list()

Function to return a named list

plot(<cumcount>)

Plots a panel of (possibly overlayed) cumulative count functions or cumulative count differences and confidence intervals.

plot_boot_cumrisk()

Plots a panel of (possibly overlayed) cumulative risk functions or cumulative risk differences and confidence intervals.

predict(<survfit>)

Predict probabilities from a survfit object

print(<result>)

Provides a print method for result objects

re_estimate(<cumcount>)

Re-estimates a result object after elements of the model_specifaction have been modified.

re_estimate(<cumrisk>)

Re-estimates a result object after elements of the model_specifaction have been modified.

re_estimate(<no_outcome>)

Re-estimates a no_outcome object after elements of the model_specifaction have been modified.

sim()

Iterates a simulation function design to assess bias across different boostrap procedures

sim_eval()

Evaluates the result of a simulation study

sim_func1()

Simulates a simple data structure with potential confounding from two variables, dependent censoring, and a competing risk

smd()

Compute a standardized mean difference for two groups

stack_cumcount()

Stack cumulative count curves from each treatment group in a results object (or list of objects) in a matrix

stack_cumcount_boot()

Stack bootstrapped cumulative count curves from each treatment group in a results object (or list of objects) in a matrix

stack_cumrisk()

Stack cumulative risk curves from each treatment group in a results object (or list of objects) in a matrix

stack_cumrisk_boot()

Stack bootstrapped cumulative risk curves from each treatment group in a results object (or list of objects) in a matrix

table2_helper_cumcount()

Helper function for make_table2.cumcount

table2_helper_cumrisk()

Helper function for make_table2.cumrisk

trim_ps()

Function for symmetric trimming predicted probabilites

update_count()

Updates the count information in a result object for later re-estimation

update_label()

Updates the label in a result object

wt_comp()

Computes treatment weights

Datasets

example1

Simulated example right censored data with confounding and dependant censoring.

example1_timevar

Simulated example right censored data with confounding, dependant censoring, and time-varying covariates.

example2

Simulated example data for ipwrisk

example3

Simulated example data for demonstration of the IPMW functionality of causalRisk.