CoxAIPW function now includes an additional crossFit parameter, which can be turned to FALSE to stop cross-fitting when all nuisance function estimators are root-n.
CoxAIPW function now includes an additional augmentation parameter in replacement of the previous ‘random_censoring’ parameter. Augmentation with respect to either propensity score, censoring or both can be selected.
Under non-proportional hazard model, CoxAIPW estimates a causal estimated that is a weighted average of the time-varying log hazard ratio. It also outputs x and y values for producing a smoothed plot of the time-varying log hazard ratio.
Update reference articles in the DESCRIPTION.
CoxAIPW function now also outputs cumulative baseline hazard function and survival function for both groups, all of which evaluated at each follow up time.
CoxAIPW function now supports an additional ‘random_censoring’ parameter, which allows simplified estimation under the random censoring assumption instead of the informative censoring assumption.
CoxAIPW function now supports data with tied event times.
Update referenced articles in the DESCRIPTION.
NEWS.md file to track changes to the
package.