## 0.39.0-0.39.02

1. make_table1 rewritten to utilize the gt package, thus allowing far more customization and exporting ability.

## 0.38.2

1. Minor updates to remove deprecated functions

## 0.38.1

1. Cleaned up documentation for compare_fits and allow user to choose alpha level for confidence intervals.

## 0.38.0

1. Improved performance of estimate_aipwrisk when fully_efficient = TRUE, by rewriting the internal function integrator and helpers.

## 0.37.25-0.37.26

1. Added compare_protocols functionality.

## 0.37.23-0.37.24

1. Added compare_fits function, which allows for the comparison of multiple fit objects

## 0.37.20-0.37.21

1. Fixed bug that caused start>stop time warnings when using time-varying covariates and times argument is provided.

## 0.37.19

1. Allow censoring and outcome models to differ by level of treatment by passing them as a list.

## 0.37.18

1. Fixed bug that prevented event and censoring ties from being broken correctly when subjects occur in multiple treatment groups and have longitudinal data.

## 0.37.17

1. Fixed bug that led to incorrect outcome times in estimate_aipwrisk when competing risks are present. Bug was introduced in version 0.36.17.

## 0.37.16

1. Updated documents to reflect behavior of SMR, normalized, and stablized weights as they are implemented in the package.

## 0.37.15

1. Fixed bug that led to NaN risk estimates when no censoring model is specified, there are tied event times, the last observed event/censoring times are tied, and everyone either has the event or censoring.

## 0.37.14

1. Fixed default behavior in forest_plot.cumrisk so that when no risk time is passed, it uses the maximum time in each input object.

## 0.37.12, 0.37.13

1. Added parameter smd_lower_limit to make_table1 that causes SMDs lower than 0.01 to be displayed as “<0.01”.

## 0.37.11

1. Fixed bug in make_table2 that caused errors when using the function within a subroutine.

## 0.37.10

1. Fixed bug that caused error when using PS trimming when missing data are present.

## 0.37.9

1. Fixed bug in make_table1 that caused error when missing data were present.

## 0.37.8

1. Expanded the color palettes for plot.cumrisk and plot.cumcount to allow any number of treatment levels.

## 0.37.7

1. Added the option to scale the rate without scaling the person-time using the scale_pt_display argument in make_table2.

## 0.37.6

1. Added the option to calculate confidence intervals for rates with the calc_rate_ci argument in make_table2.

## 0.37.5

1. Fixing bug in make_table2 that prevented selecting correct treatment reference level when multiple objects were passed.

## 0.37.3, 0.37.4

1. Exporting trim_ps function for use in causalPHR package.

## 0.37.2

1. Cleanup and optimization of efficient AIPW estimator.

## 0.37.1

1. Fixed bug that was introduced due to the change of the behavior of the right_join fuction in dplyr.

## 0.36.19

1. Bug fix. Wrong denominator for overlap weights.

## 0.36.18

1. Added overlap weights as an option.

2. Added discussion of different weighting methods to manual.

3. Bug fix. SMR weights were incorrect with more than 2 levels of treatment.

## 0.36.17

1. Changed censoring behavior. Now, censoring by end of follow-up is not treated as a censoring event to be modelled by the censoring models. To replicate analyses conducted with older versions of the package, specify tau_censor = TRUE when calling the estimation function.

2. Bug fix. When subgroup_ps is FALSE, the wrong n was being returned by the estimating functions.

3. Bug fix. When user supplied weights or a subgroup, efficient AIPW threw an error.

4. Bug fix. When missing data are present, the output prediction vector was the wrong length for AIPW.

5. Bug fix. make_table1 was inconsistent when identify_subject() was used.

6. Bug fix. When using R versions greater than 4, error was thrown when using efficient AIPW with null models.

7. Fix warning related to RNG in future package.