Fast selection of nonlinear mixed effect models using penalized likelihood

TL;DR

This paper introduces a fast, penalized likelihood-based method for selecting covariates and correlation parameters in nonlinear mixed effects models, improving efficiency for high-dimensional pharmacometric data.

Contribution

It develops a stochastic proximal gradient algorithm with adaptive step sizes and parallelization for efficient covariate selection in complex models.

Findings

The proposed method outperforms traditional grid search in speed and accuracy.

Simulation studies validate the effectiveness of the approach.

Application to pharmacokinetics data demonstrates practical utility.

Abstract

Nonlinear Mixed effects models are hidden variables models that are widely used in many fields such as pharmacometrics. In such models, the distribution characteristics of hidden variables can be specified by including several parameters such as covariates or correlations which must be selected. Recent development of pharmacogenomics has brought averaged/high dimensional problems to the field of nonlinear mixed effects modeling for which standard covariates selection techniques like stepwise methods are not well suited. The selection of covariates and correlation parameters using a penalized likelihood approach is proposed. The penalized likelihood problem is solved using a stochastic proximal gradient algorithm to avoid inner-outer iterations. Speed of convergence of the proximal gradient algorithm is improved using component-wise adaptive gradient step sizes. The practical implementation and tuning of the proximal gradient algorithm are explored using simulations. Calibration of regularization parameters is performed by minimizing the Bayesian Information Criterion using particle swarm optimization, a zero-order optimization procedure. The use of warm restart and parallelization allowed computing time to be reduced significantly . The performance of the proposed method compared to the traditional grid search strategy is explored using simulated data. Finally, an application to real data from two pharmacokinetics studies is provided, one studying an antifibrinolytic and the other studying an antibiotic.