A parsimonious personalized dose-finding model via dimension reduction.

  title={A parsimonious personalized dose-finding model via dimension reduction.},
  author={Wenzhuo Zhou and Ruoqing Zhu and Donglin Zeng},
  volume={108 3},
Learning an individualized dose rule in personalized medicine is a challenging statistical problem. Existing methods often suffer from the curse of dimensionality, especially when the decision function is estimated nonparametrically. To tackle this problem, we propose a dimension reduction framework that effectively reduces the estimation to a lower-dimensional subspace of the covariates. We exploit that the individualized dose rule can be defined in a subspace spanned by a few linear… 

Tables from this paper

A Single-Index Model With a Surface-Link for Optimizing Individualized Dose Rules
A unique contribution of this work is in the parsimonious (single-index) parametrization specifically defined for the interaction effect term.
Policy Learning for Optimal Individualized Dose Intervals
A new method to estimate an optimal dose interval, named probability dose interval (PDI), which is guaranteed better than a pre-specified threshold with a given probability (e.g., 50%).
Statistical Learning for Individualized Asset Allocation
A high-dimensional statistical learning framework for individualized asset allocation that addresses continuous-action decision-making with a large number of characteristics and shows that the individualized optimal strategy improves individual financial well-being and surpasses benchmark strategies.
A Semiparametric Instrumental Variable Approach to Optimal Treatment Regimes Under Endogeneity
A general instrumental variable approach to learning optimal treatment regimes under endogeneity is proposed and establishes identification of both value function for a given regime and optimal regimes with the aid of a binary IV, when no unmeasured confounding fails to hold.
Deep Jump Learning for Off-Policy Evaluation in Continuous Treatment Settings
A novel estimation method for OPE using deep jump learning is developed, adaptively discretizing the treatment space using deep discretization, by leveraging deep learning and multiscale change point detection.


Personalized Dose Finding Using Outcome Weighted Learning
A randomized trial design where candidate dose levels assigned to study subjects are randomly chosen from a continuous distribution within a safe range is advocated and an outcome weighted learning method based on a nonconvex loss function is proposed, which can be solved efficiently using a difference of convex functions algorithm.
A Boosting Algorithm for Estimating Generalized Propensity Scores with Continuous Treatments
This study suggests a boosting algorithm to estimate the mean function of the treatment given covariates and proposes a criterion called average absolute correlation coefficient (AACC) to determine the optimal number of trees in boosting.
Q-Learning: Flexible Learning About Useful Utilities
Simulated data results show that the GAM adapted Q-learning typically outperforms Q- Learning with linear models and other frequently-used methods based on propensity scores in terms of coverage and bias/MSE, a promising step toward a more fully general Q- learning approach to estimating optimal dynamic treatment regimes.
An Effective Semiparametric Estimation Approach for the Sufficient Dimension Reduction Model
ABSTRACT In the exploratory data analysis, the sufficient dimension reduction model has been widely used to characterize the conditional distribution of interest. Different from the existing
Greedy outcome weighted tree learning of optimal personalized treatment rules
A subgroup identification approach for inferring optimal and interpretable personalized treatment rules with high‐dimensional covariates based on a two‐step greedy tree algorithm to pursue signals in a high-dimensional space is proposed.
Estimating a sparse reduction for general regression in high dimensions
A principled method is proposed for estimating a sparse reduction, which is based on a new, revised representation of an existing well-known method called the sliced inverse regression, and a fast and efficient algorithm is developed for computing the estimator.
Semiparametric Single-Index Model for Estimating Optimal Individualized Treatment Strategy.
A new semiparametric additive single-index model for estimating individualized treatment strategy that assumes a flexible and nonparametric link function for the interaction between treatment and predictive covariates is proposed.
Individualizing drug dosage with longitudinal data
A two-step procedure to personalize drug dosage over time under the framework of a log-linear mixed-effect model that can accommodate more general time-varying covariates corresponding to random effects and is more efficient compared with existing approaches.
Doubly‐robust dynamic treatment regimen estimation via weighted least squares
A new approach is presented which builds on two established methods: Q‐learning and G‐estimation, offering the doubly robust property of the latter but with ease of implementation much more akin to the former.
The use of clinical trial data is considered in the construction of an individualized treatment rule leading to highest mean response and estimation based on l(1) penalized least squares is considered.