SBMTrees (version 1.2)

SBMTrees-package: Sequential Imputation with Bayesian Trees Mixed-Effects Models

Description

The SBMTrees package implements a Bayesian non-parametric framework for imputing missing covariates and outcomes in longitudinal data under the Missing at Random (MAR) assumption. Its core model, the Bayesian Trees Mixed-Effects Model (BMTrees), extends Mixed-Effects BART by employing centralized Dirichlet Process (CDP) Normal Mixture priors. This allows handling non-normal random effects and errors, addressing model misspecification, and capturing complex relationships.

Author

Jungang Zou <[e-mail]>

Details

SBMTrees offers tools for predicting and imputing missing values in longitudinal data using Bayesian Trees Mixed-Effects Models. The package supports various semiparametric variants, including BMTrees_R and BMTrees_RE, and integrates mixedBART as a baseline model. Key functionalities include:

- BMTrees_prediction: Predicts longitudinal outcomes based on mixed-effects models.

- sequential_imputation: Imputes missing covariates and outcomes sequentially in longitudinal datasets.

The package supports flexibility in specifying priors for random effects and errors, making it suitable for diverse longitudinal data settings. Core computations leverage efficient Gibbs samplers implemented in C++.

This package modifies and extends C++ code originally derived from the BART3 package, developed by Rodney Sparapani, which is licensed under the GNU General Public License version 2 (GPL-2).

The modified code is redistributed in accordance with the GPL-2 license. For more details on the modifications, see the package's documentation.

References

BART3 package: https://github.com/rsparapa/bnptools/tree/master, originally developed by Rodney Sparapani.

See Also

BMTrees_prediction, sequential_imputation

Examples

     ## Example of predicting longitudinal outcomes
     ## To make it faster to compile and check, we only run 30 iterations for 
     ## burn-in and 40 for posterior sampling phases.
     ## Please increase to 3000 and 4000 iterations, respectively, when running the models.
     # \donttest{
     data <- simulation_prediction(n_subject = 100, seed = 1234, nonlinear = TRUE,
             nonrandeff = TRUE, nonresidual = TRUE)
     X_train <- data$X_train
     Y_train <- data$Y_train
     Z_train <- data$Z_train
     subject_id_train <- data$subject_id_train

     X_test <- data$X_test
     Z_test <- data$Z_test
     subject_id_test <- data$subject_id_test

     model <- BMTrees_prediction(X_train, Y_train, Z_train, subject_id_train, 
              X_test, Z_test, subject_id_test, model = "BMTrees", seed = 1234)
     model$post_predictive_y_test
     
     
     
     data2 = simulation_imputation(n_subject = 100, seed = 1234, nonrandeff = TRUE, 
             nonresidual = TRUE, alligned = FALSE) 
     X_mis = data2$X_mis # get missing covariates
     Y_mis = data2$Y_mis # get missing outcomes
     Z = data2$Z # get random predictors
     subject_id = data2$subject_id  # get subject id

     model2 = sequential_imputation(X_mis, Y_mis, Z, subject_id, rep(0, 9), FALSE, 
             model = "BMTrees", nburn = 30L, npost = 40L, skip = 2L, 
             verbose = TRUE, seed = 1234)
     model2$imputed_data
     model2$imputed_data[,,10]
     # }