JQL: Jump Q-Learning for Individualized Interval-Valued Dose Rule

We provide tools to estimate the individualized interval-valued dose rule (I2DR) that maximizes the expected beneficial clinical outcome for each individual and returns an optimal interval-valued dose, by using the jump Q-learning (JQL) method. The jump Q-learning method directly models the conditional mean of the response given the dose level and the baseline covariates via jump penalized least squares regression under the framework of Q learning. We develop a searching algorithm by dynamic programming in order to find the optimal I2DR with the time complexity O(n2) and spatial complexity O(n). To alleviate the effects of misspecification of the Q-function, a residual jump Q-learning is further proposed to estimate the optimal I2DR. The outcome of interest includes the best partition of the entire dosage of interest, the regression coefficients of each partition, and the value function under the estimated I2DR as well as the Wald-type confidence interval of value function constructed through the Bootstrap.

Version: 3.6.9
Depends: caret, pdist, stats, randomForest
Published: 2019-11-15
DOI: 10.32614/CRAN.package.JQL
Author: Hengrui Cai, Chengchun shi, Rui Song, Wenbin Lu
Maintainer: Hengrui Cai <hcai5 at ncsu.edu>
License: LGPL-3
NeedsCompilation: no
CRAN checks: JQL results


Reference manual: JQL.pdf


Package source: JQL_3.6.9.tar.gz
Windows binaries: r-devel: JQL_3.6.9.zip, r-release: JQL_3.6.9.zip, r-oldrel: JQL_3.6.9.zip
macOS binaries: r-release (arm64): JQL_3.6.9.tgz, r-oldrel (arm64): JQL_3.6.9.tgz, r-release (x86_64): JQL_3.6.9.tgz, r-oldrel (x86_64): JQL_3.6.9.tgz
Old sources: JQL archive


Please use the canonical form https://CRAN.R-project.org/package=JQL to link to this page.