KMAN Publication Inc. (KMANPUB) Journal of Adolescent and Youth Psychological Studies (JAYPS) 2981-2526 7 Serial Number 44 2026 04 10 Modeling Adolescent Aggressive Behavior Using Trait Anger, Executive Function Inhibition Scores, Family Chaos, and Social Rejection via Gradient Boosted Trees Modeling Adolescent Aggressive Behavior Using Trait Anger, Executive Function Inhibition Scores, Family Chaos, and Social Rejection via Gradient Boosted Trees 1 12 10.61838/kman.jayps.5244 EN Journal Article 2025 10 19 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> The objective of this study was to rigorously predict and model adolescent aggressive behavior through the complex, non-linear interactions of trait anger, executive function inhibition scores, family chaos, and social rejection via a Gradient Boosted Trees approach.</p> <p><strong>Methods and Materials:</strong> A cross-sectional predictive study design was utilized, involving a sample of adolescents (aged – &nbsp;years) recruited from central Mexico. Data were collected using the Buss-Perry Aggression Questionnaire, State-Trait Anger Expression Inventory, a computerized Stroop Color and Word Test, the Confusion, Hubbub, and Order Scale, and the Ostracism Experience Scale. Data analysis was conducted using the XGBoost machine learning algorithm, with the dataset split into an training set ( ) and a testing set ( ). Model evaluation and feature importance were calculated using , , , and SHapley Additive exPlanations (SHAP) values.</p> <p><strong>Findings:</strong> The XGBoost model exhibited robust predictive performance on the unseen testing set ( , , ). Based on Mean Absolute SHAP values, trait anger was identified as the paramount predictor of aggressive behavior ( &nbsp;relative importance), sequentially followed by social rejection ( ), family chaos ( ), and executive function inhibition ( ). Furthermore, interaction analysis revealed profound synergistic effects, with the strongest non-linear interaction occurring between trait anger and executive function inhibition (SHAP Interaction Value ), alongside a highly significant interaction between social rejection and family chaos (SHAP Interaction Value ).</p> <p><strong>Conclusion:</strong> Adolescent aggression is a non-linear, compounded phenomenon that is most accurately predicted by the synergistic interaction of high emotional reactivity, poor cognitive inhibition, and severe environmental stressors.</p> </td> </tr> </tbody> </table> https://journals.kmanpub.com/index.php/jayps/article/download/5244/9522