Enhancing Forecast Accuracy for Damped Trend and Multiplicative Seasonality: A Hybrid Time-Series Framework with Simulation Study.
Forecasting time series characterized by a damped trend and multiplicative seasonality (DTMS) presents a significant challenge, as conventional models like ARIMA and ETS often fail to capture their complex, non-linear interactions. This study bridges this gap by developing and evaluating a novel hyb...
Gardado en:
| Main Authors: | , , , |
|---|---|
| Formato: | Online |
| Idioma: | inglés |
| Publicado: |
Journal of Science
2026
|
| Subjects: | |
| Acceso en liña: | http://journals.juniv.edu/index.php/jos/article/view/386 |
| Tags: |
Engadir etiqueta
Sen Etiquetas, Sexa o primeiro en etiquetar este rexistro!
|
| _version_ | 1873958350154104832 |
|---|---|
| author | Md. Kamrul Hasan Amrin Binte Ahmed K. M. Zahidul Islam Rumana Rois |
| author_facet | Md. Kamrul Hasan Amrin Binte Ahmed K. M. Zahidul Islam Rumana Rois |
| author_sort | Md. Kamrul Hasan |
| collection | OJS |
| description | Forecasting time series characterized by a damped trend and multiplicative seasonality (DTMS) presents a significant challenge, as conventional models like ARIMA and ETS often fail to capture their complex, non-linear interactions. This study bridges this gap by developing and evaluating a novel hybrid forecasting framework that synergistically combines statistical and machine learning (ML) approaches. We hypothesize that while statistical models capture linear components, ML models excel at modeling non-linear residuals. Our methodology employs a comprehensive simulation study to systematically control data characteristics, alongside an empirical analysis of a real financial dataset from the Dhaka Stock Exchange. Results demonstrate that hybrid models, particularly SVR-ANN and SVR-ETS, significantly outperform individual and other hybrid models across all accuracy metrics (RMSE, MAE, MAPE, MASE). The simulation confirmed this superiority, with the top hybrids achieving a mean MASE of 0.701. This indicates that an ML model, especially SVR, is highly effective in modeling the complex residuals from a primary statistical forecast. We conclude that a hybrid paradigm integrating statistical and ML techniques offers a robust and superior solution for forecasting DTMS series, providing practitioners with evidence-based guidance for enhanced accuracy. |
| format | Online |
| id | ojs-386 |
| institution | My University |
| language | eng |
| publishDate | 2026 |
| publisher | Journal of Science |
| record_format | ojs |
| spelling | ojs-3862026-04-01T06:14:41Z Enhancing Forecast Accuracy for Damped Trend and Multiplicative Seasonality: A Hybrid Time-Series Framework with Simulation Study. Md. Kamrul Hasan Amrin Binte Ahmed K. M. Zahidul Islam Rumana Rois Dhaka Stock Exchange, Damped trend and multiplicative seasonality, Hybrid machine learning, Low-volatility regime; Support Vector Regression. Forecasting time series characterized by a damped trend and multiplicative seasonality (DTMS) presents a significant challenge, as conventional models like ARIMA and ETS often fail to capture their complex, non-linear interactions. This study bridges this gap by developing and evaluating a novel hybrid forecasting framework that synergistically combines statistical and machine learning (ML) approaches. We hypothesize that while statistical models capture linear components, ML models excel at modeling non-linear residuals. Our methodology employs a comprehensive simulation study to systematically control data characteristics, alongside an empirical analysis of a real financial dataset from the Dhaka Stock Exchange. Results demonstrate that hybrid models, particularly SVR-ANN and SVR-ETS, significantly outperform individual and other hybrid models across all accuracy metrics (RMSE, MAE, MAPE, MASE). The simulation confirmed this superiority, with the top hybrids achieving a mean MASE of 0.701. This indicates that an ML model, especially SVR, is highly effective in modeling the complex residuals from a primary statistical forecast. We conclude that a hybrid paradigm integrating statistical and ML techniques offers a robust and superior solution for forecasting DTMS series, providing practitioners with evidence-based guidance for enhanced accuracy. Journal of Science 2026-01-13 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion Peer-reviewed Conference Proceedings Submission application/pdf http://journals.juniv.edu/index.php/jos/article/view/386 10.59185/jos.v45i1.386 Journal of Science; Vol. 45 No. 1 (2025): Jahangirnagar University Journal of Science 10.59185/jos.v45i1 eng http://journals.juniv.edu/index.php/jos/article/view/386/318 Copyright (c) 2026 Jahangirnagar University Journal of Science |
| spellingShingle | Dhaka Stock Exchange, Damped trend and multiplicative seasonality, Hybrid machine learning, Low-volatility regime; Support Vector Regression. Md. Kamrul Hasan Amrin Binte Ahmed K. M. Zahidul Islam Rumana Rois Enhancing Forecast Accuracy for Damped Trend and Multiplicative Seasonality: A Hybrid Time-Series Framework with Simulation Study. |
| title | Enhancing Forecast Accuracy for Damped Trend and Multiplicative Seasonality: A Hybrid Time-Series Framework with Simulation Study. |
| title_full | Enhancing Forecast Accuracy for Damped Trend and Multiplicative Seasonality: A Hybrid Time-Series Framework with Simulation Study. |
| title_fullStr | Enhancing Forecast Accuracy for Damped Trend and Multiplicative Seasonality: A Hybrid Time-Series Framework with Simulation Study. |
| title_full_unstemmed | Enhancing Forecast Accuracy for Damped Trend and Multiplicative Seasonality: A Hybrid Time-Series Framework with Simulation Study. |
| title_short | Enhancing Forecast Accuracy for Damped Trend and Multiplicative Seasonality: A Hybrid Time-Series Framework with Simulation Study. |
| title_sort | enhancing forecast accuracy for damped trend and multiplicative seasonality a hybrid time series framework with simulation study |
| topic | Dhaka Stock Exchange, Damped trend and multiplicative seasonality, Hybrid machine learning, Low-volatility regime; Support Vector Regression. |
| topic_facet | Dhaka Stock Exchange, Damped trend and multiplicative seasonality, Hybrid machine learning, Low-volatility regime; Support Vector Regression. |
| url | http://journals.juniv.edu/index.php/jos/article/view/386 |
| work_keys_str_mv | AT mdkamrulhasan enhancingforecastaccuracyfordampedtrendandmultiplicativeseasonalityahybridtimeseriesframeworkwithsimulationstudy AT amrinbinteahmed enhancingforecastaccuracyfordampedtrendandmultiplicativeseasonalityahybridtimeseriesframeworkwithsimulationstudy AT kmzahidulislam enhancingforecastaccuracyfordampedtrendandmultiplicativeseasonalityahybridtimeseriesframeworkwithsimulationstudy AT rumanarois enhancingforecastaccuracyfordampedtrendandmultiplicativeseasonalityahybridtimeseriesframeworkwithsimulationstudy |