HSML-CP: A Hybrid Statistical–Machine Learning Framework with  Conformal Prediction for Uncertainty-Aware Forecasting in Inventory  Optimization and Financial Risk Management

Suryasivaji Killi

HSML-CP: A Hybrid Statistical–Machine Learning Framework with Conformal Prediction for Uncertainty-Aware Forecasting in Inventory Optimization and Financial Risk Management

Authors

Suryasivaji Killi

Keywords:

Hybrid Forecasting, Conformal Prediction, ARIMA, Uncertainty Quantification, Adaptive Weight Optimisation, Inventory Optimization, Demand Forecasting, Financial Time Series, Supply Chain AI, Micro-Fulfillment

Abstract

Accurate uncertainty-aware forecasting is important for capital-efficient inventory and financialoptimization of AI-based micro-fulfillment platforms. Classical statistical time series models, such asARIMA (Autoregressive Integrated Moving Average), yield interpretable forecasts but struggle to capture nonlinear dependencies. Traditional machine learning approaches

References

George E. P. Box, Gwilym M. Jenkins, Gregory C. Reinsel, and Greta M. Ljung, Time Series Analysis: Forecasting and Control, 5th ed. John Wiley & Sons, 2015. Available: https://www.wiley.com/enus/Time+Series+Analysis

Downloads

Published

2026-06-16

How to Cite

Suryasivaji Killi. (2026). HSML-CP: A Hybrid Statistical–Machine Learning Framework with Conformal Prediction for Uncertainty-Aware Forecasting in Inventory Optimization and Financial Risk Management . Journal of Computational Analysis and Applications (JoCAAA), 35(6), 146–158. Retrieved from https://www.eudoxuspress.com/index.php/pub/article/view/5587