boostmtree - Boosted multivariate trees for longitudinal data.

Implements Friedman's gradient descent boosting algorithm for modeling longitudinal response using multivariate tree base learners. Longitudinal response could be continuous, binary, nominal or ordinal. A time-covariate interaction effect is modeled using penalized B-splines (P-splines) with estimated adaptive smoothing parameter. Although the package is design for longitudinal data, it can handle cross-sectional data as well.

Package Overview

This package contains many useful functions and users should read the help file in its entirety for details. However, we briefly mention several key functions that may make it easier to navigate and understand the layout of the package.

1. boostmtree - This is the main entry point to the package. It grows a multivariate tree using user supplied training data. Trees are grown using the randomForestSRC R-package.

2. predict.boostmtree - Used for prediction. Predicted values are obtained by dropping the user supplied test data down the grow forest. The resulting object has class (rfsrc, predict).

Authors

Hemant Ishwaran, Amol Pande and Udaya B. Kogalur

References

Friedman J.H. (2001). Greedy function approximation: a gradient boosting machine, Ann. of Statist., 5:1189-1232.

Pande A., Li L., Rajeswaran J., Ehrlinger J., Kogalur U., Blackstone E.H., Ishwaran H. Boosted Multivariate Trees for Longitudinal Data Mach Learn, . 2017 Feb;106(2):277-305. doi: 10.1007/s10994-016-5597-1.