Changes in version 1.5.5 New Features Bug Fixes - Fixes bug where clustered cross-fitting folds are silently disabled and the package returns row-level V-fold CV instead (see issue #188). - TMLE fluctutation step was using training data and not validation data resulting in EIF not being mean zero. - fits_r/fits_m changed to fits_treatment/fits_outcome. General - Removed the lmtp_control() option return_full_fits. A summary data.table is now always returned. Changes in version 1.5.4 (2026-05-07) New Features Bug Fixes General - Added check that the number of Super Learner cross-validation folds isn't greater than the number of clusters (see issue #105). - origami has been removed as a dependency. - Now requires ife version be greater than 0.1.2 (see issue #180). - Risk ratio and odds ratio in lmtp_contrast are now estimated on the log scale and then the estimate and confidence intervals are exponentiated (see issue #180). Changes in version 1.5.3 (2025-07-24) New Features Bug Fixes - Fixes bug where the shifted argument wouldn't work with multivariate exposure (see issue #175). - Fixes bug where mtp = FALSE wouldn't work with multivariate exposure. General Changes in version 1.5.2 (2025-06-13) New Features Bug Fixes General - The default for the mtp argument has been changed from FALSE to TRUE (see issue #170). - A warning message is now printed in lmtp_contrast() that p-values aren't adjusted for multiple comparisons (see issue #172). - A warning message is now printed in lmtp_contrast() if ref is a constant. Changes in version 1.5.1 (2025-05-20) New Features Bug Fixes - Super learner prediction failing for some learners. Now using onlySL = TRUE in predict.SuperLearner (see issue #162). General - Making sure to only pass the necessary variables to predict.SuperLearner to suppress some warnings. Changes in version 1.5.0 (2025-05-01) New Features - Added the ability to estimate the so-called "total effect" for survival outcome with competing risks (see issue #143). - IPW and g-computation estimators are no-longer supported; they have been given deprecation errors. - Added lmtp_survival() function for estimating the entire survival curve. Enforces monotonicity using isotonic regression (see issue #140). Bug Fixes - Using fitted values from isotonic regression in lmtp_survival() instead of the original values (see issue #149). General - Removed dependency on schoolmath which used a very slow function for testing if a vector was "decimalish". - Fixed "F used instead of FALSE" error in CRANs tests. Changes in version 1.4.0 (2024-06-27) New Features - Can now estimate the effects of simultaneous interventions on multiple variables. - New pre-packaged shift function, ipsi() for estimating IPSI effects using the risk ratio. - lmtp_control() now replaces extra estimator arguments. Bug Fixes - Standard errors now incorporate survey weights (see issue #134). - Bug fix when shift is NULL and data is a tibble (see issue #137) General - The intervention_type argument has been fully deprecated. - Now attempting to detect intervention type errors (see issue #98). Changes in version 1.3.3 (2024-03-26) New Features Bug Fixes - Fixed a bug where estimators return incorrect parameter estimates for a specific DGP (see issue #130) General Changes in version 1.3.2 (2023-07-19) New Features Bug Fixes - Fixed bug in calculation of EIF where density ratios were not non-cumulative product ratios. Previous variance estimates starting with version 1.0 were incorrect. Point-estimates remain unaffected. General - Updating citations Changes in version 1.3.1 (2022-09-07) New Features - Added parameter .return_full_fits. Allows the user to decide if full SuperLearner fit should be returned (issue #119). - intervention_type argument replaced with mtp. Bug Fixes - Added a check for fits$id being NULL. Fixes a backwards compatibility bug (issue #117). - data.table version must be 1.13.0 or later. This was when the function fcase was released (issue #122). General - Changed 'effect' to 'estimate' in 'Population mean effects' portion of output (issue #120). Changes in version 1.3.0 (2022-05-21) New Features Bug Fixes General - Major internal refactor. Argument checking is now performed using checkmate package. - .SL_folds argument split into .learners_outcome_folds and .learners_trt_folds. Changes in version 1.1.0 New Features Bug Fixes - Corrected standard errors when providing id with lmtp_contrast (issue #110). General - Removed the requirement that folds must be greater than 1 (issue #112). Changes in version 1.0.0 (2021-09-29) New Features - New shifted parameter for directly passing shifted data instead of using a shift function (issue #89). - New intervention_type parameter required for specifying if the intervention of interest is a static regime, a dynamic regime, or a modified treatment policy (issue #94). - return_all_ratios removed as an argument. Returned density ratios are now non-cumulative product ratios. Bug Fixes - Density ratio trimming now occurs in the same spot for all estimators and is only performed on non-cumulative product ratios (issue #\93). - Fixed issue where lmtp_tmle and lmtp_sdr weren't using validation set density ratios. - No longer fails when data is a data.table (issue #88). General - Removing extra column in sim_point_surv data set (issue #91). - Paper citation updated with release in JASA (issue #103). Changes in version 0.9.1 (2021-08-18) Bug Fixes - Fixed a bug that caused failure when knitting the getting-started.Rmd vignette when using new version of the future package (issue #100). General - GitHub links added to DESCRIPTION (issue # 99). Bug Fixes - Fixed a bug that caused failure when no variation existed in the outcome at a type point (issue #92). - No longer fails when data is a data.table (issue #88). General - Removing extra column in sim_point_surv data set (issue #91). Changes in version 0.9.0 (2021-02-22) New Features - New weights parameter for observation sampling weights (issue #78). - For time-to-event analysis, survival probability is now estimated instead of the cumulative incidence. This fixes a bug with IPW and survival problems. - Outcome type now accepts "survival" for explicit indication of a survival outcome (issue #76). Because of this lmtp_ipw() now requires setting the outcome type. - New .trimming parameter for trimming extreme density ratios. - New .SL_folds parameter that controls the splits used for fitting the SuperLearner (issue #84). - New .return_all_ratios parameter that allows for returning non-cumulative product density ratios to the user. - bound parameter renamed to .bound. Bug Fixes - Fixed a bug that caused the final estimate to be incorrectly estimated with SDR (issue #87). - Fixed a bug that outputted outcome regressions and density ratios in incorrect order compared to the original data. - Fixed a bug in the missing data check that threw an error for missing data after an observation experiences the outcome. - Fixed a bug in the calculation of standard errors when the id parameter is specified. - Fixed a bug that resulted in NA censoring indicators throwing an error for missing data. - Fixed a bug about values() being deprecated in the future package (issue #82). - Fixed a warning from the future package regarding random number generation (issue #81). - Fixed create_node_list() returns description (issue #77). Dependencies - slider dependency removed. - data.table added as a dependency. General - event_locf() speed greatly improved (issue #80). - Migrated continuous integration from Travis-CI to GitHub Actions. - Added a NEWS.md file to track changes to the package. - License change to GPL-3.