AFMF-SFBM-A09: The Alaali Behavioral Sustainability Forecasting Artificial Intelligence Model (A-BSF-AI): Development and Comparative Evaluation Through Synthetic Simulations

Authors

DOI:

https://doi.org/10.31098/lomr.v5i2.4859

Keywords:

Behavioral Sustainability, Adaptive Forecasting, Artificial Intelligence, A-BSF-AI, AFMF, DFAS, Synthetic Simulation.

Abstract

This study develops the Alaali Behavioral Sustainability Forecasting Artificial Intelligence Model (A-BSF-AI; AFMF-SFBM-A09) as an adaptive extension of the Alaali Behavioral Sustainability Forecasting Model (A-BSF) within the Alaali Financial Models Meta-Framework (AFMF). It evaluates whether sequential coefficient updating improves forecasting accuracy relative to fixed and alternative adaptive estimation methods under controlled synthetic conditions. The experiment comprises 30 independently seeded simulation runs, each containing 15,000 observations across 10 temporal periods, producing 450,000 observations. The data-generating process incorporates behavioral, environmental, social, and governance (ESG), market, cyclical, and time-dependent influences. Five forecasting specifications are compared: restricted ordinary least squares (OLS), correctly specified OLS, rolling OLS, recursive least squares (RLS), and adaptive clipped normalized least mean squares (NLMS). Chronological partitioning allocates 70% of observations to training, 15% to validation, and 15% to testing. Forecasting performance is assessed using root mean squared error (RMSE) and paired statistical comparisons. Adaptive NLMS achieves a mean testing RMSE of 0.120378, compared with 0.138282 for restricted OLS, representing a 12.94% improvement (p < 0.001). It also outperforms the tested RLS specification (p < 0.001). However, correctly specified OLS achieves the lowest mean RMSE (0.119822), while rolling OLS performs comparably to adaptive NLMS (p = 0.118). These findings demonstrate a conditional forecasting advantage for sequential adaptation rather than universal superiority. The study additionally proposes the Behavioral Intelligence Coefficient (BIC), Ethical Responsibility Index (ERI), and Intelligent Maturity Index (IMI) as conceptual indicators requiring separate operational validation. Its contribution lies in developing the A-BSF-AI architecture and providing a reproducible comparative computational assessment of adaptive behavioral sustainability forecasting within AFMF and Dynamic Financial Applied Meta-Science (DFAS).

Downloads

Published

2026-11-20

Citation Check

How to Cite

Alaali, H. M. H. (2026). AFMF-SFBM-A09: The Alaali Behavioral Sustainability Forecasting Artificial Intelligence Model (A-BSF-AI): Development and Comparative Evaluation Through Synthetic Simulations. Logistic and Operation Management Research (LOMR) , 5(2), 40–63. https://doi.org/10.31098/lomr.v5i2.4859