Journal ArticleNovel methods for adaptive time-series forecasting and prediction-interval constructionAmaratunga, Dhammika; Cabrera, Javier; Diaz-Tena, Nuria; Katehakis, Michael N; Lin, Chun-Pang; Wang, Jin (ORCID:0000000340474297)<title>Abstract</title> <p>We propose novel methods for adaptive series forecasting and prediction-interval construction, illustrated with COVID-19 case and death counts. Our framework applies an automated transformation to reduce heteroscedasticity, then imposes a constrained smoothing near the forecast edge via robust quadratic regression, emphasizing recent data. A Long Short-Term Memory (LSTM) model combined with ARIMA-based noise correction further refines the forecast. Compared to conventional methods (e.g., ARIMA alone, unprocessed deep learning), this adaptive approach achieves superior metrics and reliable bootstrap-derived confidence and prediction intervals. We also highlight how reinforcement learning (RL) can offer promising avenues for real-time decision-making and further improvements in forecasting adaptability.</p>Springer2025-09-0110669719Annals of Operations Research0254-5330https://doi.org/10.1007/s10479-025-06795-21662629Adaptive learning, Pre-processing, Forecast, Confidence and prediction intervalNational Science Foundation