Randomized Deep Hopfield Network with Multiple Output Layers for Volatility Time Series Forecasting

<p>Volatility forecasting plays a critical role in risk management and financial decision-making by facilitating the prediction of market fluctuations. However, the inherent complexity and irregular variations in volatility time series data present significant challenges for accurate modeling....

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Aryan Bhambu (18767731) (author)
مؤلفون آخرون: Selvaraju Natarajan (20884244) (author), Ponnuthurai Nagaratnam Suganthan (11274636) (author)
منشور في: 2025
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513534140874752
author Aryan Bhambu (18767731)
author2 Selvaraju Natarajan (20884244)
Ponnuthurai Nagaratnam Suganthan (11274636)
author2_role author
author
author_facet Aryan Bhambu (18767731)
Selvaraju Natarajan (20884244)
Ponnuthurai Nagaratnam Suganthan (11274636)
author_role author
dc.creator.none.fl_str_mv Aryan Bhambu (18767731)
Selvaraju Natarajan (20884244)
Ponnuthurai Nagaratnam Suganthan (11274636)
dc.date.none.fl_str_mv 2025-10-13T09:00:00Z
dc.identifier.none.fl_str_mv 10.1016/j.neunet.2025.108207
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Randomized_Deep_Hopfield_Network_with_Multiple_Output_Layers_for_Volatility_Time_Series_Forecasting/30365764
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Information and computing sciences
Artificial intelligence
Machine learning
Software engineering
Forecasting
Hopfield network
Volatility
Multiple output layers
Ensemble Deep Random Vector Functional Link
Time Series
Transformers
Ensemble Learning
Deep Neural Networks
dc.title.none.fl_str_mv Randomized Deep Hopfield Network with Multiple Output Layers for Volatility Time Series Forecasting
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p>Volatility forecasting plays a critical role in risk management and financial decision-making by facilitating the prediction of market fluctuations. However, the inherent complexity and irregular variations in volatility time series data present significant challenges for accurate modeling. This study proposes a novel ensemble deep randomized Hopfield network (edRHN) for volatility forecasting. The proposed network incorporates multiple stacked hidden layers with randomly generated parameters within a predefined range, which remains fixed while the output weights are determined using a closed-form solution. Each hidden layer representation contributes to training an output layer, and the aggregation of these output layers forms the final output. Technical indicators are used to identify market trends to allow more informed decision-making. The neuron pruning strategy and feature analysis are further used to eliminate noisy information and less relevant features from randomly generated features, optimizing network efficiency and performance. A comprehensive comparative study was conducted against various state-of-the-art models across ten diverse volatility time-series datasets. The experimental results highlight the superior predictive performance of the proposed model, as demonstrated by three error metrics and statistical tests.</p><h2>Other Information</h2> <p> Published in: Neural Networks<br> License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.neunet.2025.108207" target="_blank">https://dx.doi.org/10.1016/j.neunet.2025.108207</a></p>
eu_rights_str_mv openAccess
id Manara2_c7843d96c2d454e91f9f9cd21be29219
identifier_str_mv 10.1016/j.neunet.2025.108207
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/30365764
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Randomized Deep Hopfield Network with Multiple Output Layers for Volatility Time Series ForecastingAryan Bhambu (18767731)Selvaraju Natarajan (20884244)Ponnuthurai Nagaratnam Suganthan (11274636)Information and computing sciencesArtificial intelligenceMachine learningSoftware engineeringForecastingHopfield networkVolatilityMultiple output layersEnsemble Deep Random Vector Functional LinkTime SeriesTransformersEnsemble LearningDeep Neural Networks<p>Volatility forecasting plays a critical role in risk management and financial decision-making by facilitating the prediction of market fluctuations. However, the inherent complexity and irregular variations in volatility time series data present significant challenges for accurate modeling. This study proposes a novel ensemble deep randomized Hopfield network (edRHN) for volatility forecasting. The proposed network incorporates multiple stacked hidden layers with randomly generated parameters within a predefined range, which remains fixed while the output weights are determined using a closed-form solution. Each hidden layer representation contributes to training an output layer, and the aggregation of these output layers forms the final output. Technical indicators are used to identify market trends to allow more informed decision-making. The neuron pruning strategy and feature analysis are further used to eliminate noisy information and less relevant features from randomly generated features, optimizing network efficiency and performance. A comprehensive comparative study was conducted against various state-of-the-art models across ten diverse volatility time-series datasets. The experimental results highlight the superior predictive performance of the proposed model, as demonstrated by three error metrics and statistical tests.</p><h2>Other Information</h2> <p> Published in: Neural Networks<br> License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.neunet.2025.108207" target="_blank">https://dx.doi.org/10.1016/j.neunet.2025.108207</a></p>2025-10-13T09:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1016/j.neunet.2025.108207https://figshare.com/articles/journal_contribution/Randomized_Deep_Hopfield_Network_with_Multiple_Output_Layers_for_Volatility_Time_Series_Forecasting/30365764CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/303657642025-10-13T09:00:00Z
spellingShingle Randomized Deep Hopfield Network with Multiple Output Layers for Volatility Time Series Forecasting
Aryan Bhambu (18767731)
Information and computing sciences
Artificial intelligence
Machine learning
Software engineering
Forecasting
Hopfield network
Volatility
Multiple output layers
Ensemble Deep Random Vector Functional Link
Time Series
Transformers
Ensemble Learning
Deep Neural Networks
status_str publishedVersion
title Randomized Deep Hopfield Network with Multiple Output Layers for Volatility Time Series Forecasting
title_full Randomized Deep Hopfield Network with Multiple Output Layers for Volatility Time Series Forecasting
title_fullStr Randomized Deep Hopfield Network with Multiple Output Layers for Volatility Time Series Forecasting
title_full_unstemmed Randomized Deep Hopfield Network with Multiple Output Layers for Volatility Time Series Forecasting
title_short Randomized Deep Hopfield Network with Multiple Output Layers for Volatility Time Series Forecasting
title_sort Randomized Deep Hopfield Network with Multiple Output Layers for Volatility Time Series Forecasting
topic Information and computing sciences
Artificial intelligence
Machine learning
Software engineering
Forecasting
Hopfield network
Volatility
Multiple output layers
Ensemble Deep Random Vector Functional Link
Time Series
Transformers
Ensemble Learning
Deep Neural Networks