New Longitudinal Random Forest framework handles sparse, irregular data

Researchers have introduced a new Longitudinal Random Forest (LRF) framework designed to analyze sparse and irregular longitudinal data. This framework captures individual response trajectories while accounting for within-node correlation and between-node heterogeneity. It offers two variants, LRF-PACE and LRF-adaptiveLMM, which utilize nonparametric and semiparametric smoothers, respectively, and provides methods for interpreting covariate effects and predicting future trajectories. Simulations show LRF outperforms existing methods, particularly under severe data sparsity. AI

IMPACT This new framework could improve the analysis of complex biological and clinical data by better capturing individual patient trajectories.

RANK_REASON The item describes a new methodology published in an arXiv preprint. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv stat.ML →

• adaptive linear mixed-effects models
• Longitudinal Random Forest
• LRF-adaptiveLMM
• LRF-PACE
• Principal Analysis by Conditional Expectation
• trajectory-based permutation variable importance measure

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →