{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T13:40:48Z","timestamp":1792762848768,"version":"4.5.0"},"posted":{"date-parts":[[2026]]},"group-title":"SSRN","reference-count":57,"publisher":"Elsevier BV","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>We investigate how temporal dependencies, path signatures, and cross-sectional commonality can be leveraged to improve intraday realized volatility forecasting. To this end, we develop two recurrent neural network architectures: (i) a multi-scale attention mechanism that captures horizon-dependent dynamics, and (ii) a signature-augmented model that encodes the geometric features of high-frequency price paths. In addition, we introduce a cluster-based training scheme that pools stocks within sectors and conditions on sector-level realized volatility. Using one-minute data for large-cap Chinese A-share stocks, we benchmark our models against standard econometric methods, tree-based learners, and conventional recurrent networks. Our results show that nonlinear models consistently outperform linear specifications, that multi-scale attention and signature augmentation yield systematic gains across short and long horizons, and that cluster-based training delivers the largest and most robust improvements. Diebold-Mariano tests and model confidence set procedures confirm a clear hierarchy among pooling strategies, with sector-cluster-based models outperforming market-augmented, fully pooled, and stock-specific approaches. These findings highlight sector-level comovement as a persistent and exploitable source of predictive power in intraday volatility.<\/jats:p>","DOI":"10.2139\/ssrn.6287198","type":"posted-content","created":{"date-parts":[[2026,3,12]],"date-time":"2026-03-12T18:41:46Z","timestamp":1773340906000},"source":"Crossref","is-referenced-by-count":0,"title":["Multi-scale and Path Signature-informed Deep Learning for Intraday Volatility Forecasting"],"prefix":"10.2139","author":[{"given":"Ruolan","family":"Ouyang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1270-7313","authenticated-orcid":true,"given":"Jie","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3463-6774","authenticated-orcid":true,"given":"Yihan","family":"Zou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6917-221X","authenticated-orcid":true,"given":"Yukun","family":"Shi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"issue":"1","key":"ref1","doi-asserted-by":"crossref","first-page":"43","DOI":"10.1016\/S0304-405X(01)00055-1","article-title":"The distribution of realized stock return volatility","volume":"61","author":"T G Andersen","year":"2001","journal-title":"Journal of Financial Economics"},{"issue":"2","key":"ref2","doi-asserted-by":"crossref","first-page":"579","DOI":"10.1111\/1468-0262.00418","article-title":"Modeling and forecasting realized volatility","volume":"71","author":"T G Andersen","year":"2003","journal-title":"Econometrica"},{"issue":"5","key":"ref3","doi-asserted-by":"crossref","first-page":"457","DOI":"10.1016\/S0927-5398(99)00013-4","article-title":"Forecasting financial market volatility: Sample frequency vis-a-vis forecast horizon","volume":"6","author":"T G Andersen","year":"1999","journal-title":"Journal of Empirical Finance"},{"issue":"528","key":"ref4","doi-asserted-by":"crossref","first-page":"1695","DOI":"10.1080\/01621459.2018.1512864","article-title":"Time-varying periodicity in intraday volatility","volume":"114","author":"T G Andersen","year":"2019","journal-title":"Journal of the American Statistical Association"},{"issue":"3","key":"ref5","doi-asserted-by":"crossref","first-page":"817","DOI":"10.2307\/2938229","article-title":"Heteroskedasticity and autocorrelation consistent covariance matrix estimation","volume":"59","author":"D W K Andrews","year":"1991","journal-title":"Econometrica"},{"issue":"2","key":"ref6","doi-asserted-by":"crossref","first-page":"307","DOI":"10.1016\/j.jeconom.2017.08.010","article-title":"Generalized dynamic factor models and volatilities: estimation and forecasting","volume":"201","author":"M Barigozzi","year":"2017","journal-title":"Journal of Econometrics"},{"issue":"5","key":"ref7","doi-asserted-by":"crossref","first-page":"457","DOI":"10.1002\/jae.691","article-title":"Estimating quadratic variation using realized variance","volume":"17","author":"O E Barndorff-Nielsen","year":"2002","journal-title":"Journal of Applied Econometrics"},{"issue":"1","key":"ref8","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1093\/jjfinec\/nbh001","article-title":"Power and bipower variation with stochastic volatility and jumps","volume":"2","author":"O E Barndorff-Nielsen","year":"2004","journal-title":"Journal of Financial Econometrics"},{"issue":"1","key":"ref9","doi-asserted-by":"crossref","first-page":"329","DOI":"10.1016\/j.ejor.2015.12.010","article-title":"Modeling and forecasting exchange rate volatility in time-frequency domain","volume":"251","author":"J Barunik","year":"2016","journal-title":"European Journal of Operational Research"},{"issue":"5","key":"ref10","doi-asserted-by":"crossref","first-page":"961","DOI":"10.1093\/jjfinec\/nbaa049","article-title":"Decoupling the short-and long-term behavior of stochastic volatility","volume":"20","author":"M Bennedsen","year":"2022","journal-title":"Journal of Financial Econometrics"},{"issue":"3","key":"ref11","doi-asserted-by":"crossref","first-page":"307","DOI":"10.1016\/0304-4076(86)90063-1","article-title":"Generalized autoregressive conditional heteroskedasticity","volume":"31","author":"T Bollerslev","year":"1986","journal-title":"Journal of Econometrics"},{"issue":"7","key":"ref12","doi-asserted-by":"crossref","first-page":"2729","DOI":"10.1093\/rfs\/hhy041","article-title":"Risk everywhere: modeling and managing volatility","volume":"31","author":"T Bollerslev","year":"2018","journal-title":"The Review of Financial Studies"},{"issue":"4","key":"ref13","first-page":"73","article-title":"Dilated convolutional neural networks for time series forecasting","volume":"22","author":"A Borovykh","year":"2018","journal-title":"Journal of Computational Finance"},{"issue":"3","key":"ref14","doi-asserted-by":"crossref","first-page":"563","DOI":"10.1007\/s10479-022-05014-6","article-title":"Change-points and functional features of intraday volatility in china stock market","volume":"352","author":"S Boubaker","year":"2022","journal-title":"Annals of Operations Research"},{"key":"ref15","doi-asserted-by":"crossref","DOI":"10.1016\/j.jempfin.2024.101524","article-title":"Forecasting realized volatility: does anything beat linear models","volume":"78","author":"R R Branco","year":"2024","journal-title":"Journal of Empirical Finance"},{"issue":"3","key":"ref16","doi-asserted-by":"crossref","first-page":"502","DOI":"10.1093\/jjfinec\/nbaa008","article-title":"Realized volatility forecasting with neural networks","volume":"18","author":"A Bucci","year":"2020","journal-title":"Journal of Financial Econometrics"},{"key":"ref17","doi-asserted-by":"crossref","first-page":"127","DOI":"10.1016\/j.physa.2018.11.061","article-title":"Financial time series forecasting model based on CEEMDAN and LSTM","volume":"519","author":"J Cao","year":"2019","journal-title":"Physica A: Statistical Mechanics and its Applications"},{"key":"ref18","doi-asserted-by":"crossref","first-page":"785","DOI":"10.1145\/2939672.2939785","article-title":"XGBoost: a scalable tree boosting system","author":"T Chen","year":"2016","journal-title":"Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD'16"},{"key":"ref19","doi-asserted-by":"crossref","first-page":"491","DOI":"10.1016\/j.neucom.2020.03.011","article-title":"Probabilistic forecasting with temporal convolutional neural network","volume":"399","author":"Y Chen","year":"2020","journal-title":"Neurocomputing"},{"key":"ref20","first-page":"3","article-title":"A primer on the signature method in machine learning","author":"I Chevyrev","year":"2025","journal-title":"Signature Methods in Finance: An Introduction with Computational Applications"},{"issue":"2","key":"ref21","doi-asserted-by":"crossref","first-page":"125","DOI":"10.1016\/S0304-405X(98)00034-8","article-title":"The relation between implied and realized volatility","volume":"50","author":"B J Christensen","year":"1998","journal-title":"Journal of Financial Economics"},{"issue":"5","key":"ref22","doi-asserted-by":"crossref","first-page":"1680","DOI":"10.1093\/jjfinec\/nbac020","article-title":"A machine learning approach to volatility forecasting","volume":"21","author":"K Christensen","year":"2023","journal-title":"Journal of Financial Econometrics"},{"issue":"2","key":"ref23","doi-asserted-by":"crossref","first-page":"174","DOI":"10.1093\/jjfinec\/nbp001","article-title":"A simple approximate long-memory model of realized volatility","volume":"7","author":"F Corsi","year":"2009","journal-title":"Journal of Financial Econometrics"},{"key":"ref24","doi-asserted-by":"crossref","DOI":"10.1016\/j.eswa.2023.120418","article-title":"Volatility forecasting with hybrid neural networks methods for risk parity investment strategies","volume":"229","author":"L Di Persio","year":"2023","journal-title":"Expert Systems with Applications"},{"key":"ref25","doi-asserted-by":"crossref","DOI":"10.1016\/j.irfa.2022.102255","article-title":"Futures volatility forecasting based on big data analytics with incorporating an order imbalance effect","volume":"83","author":"S Ding","year":"2022","journal-title":"International Review of Financial Analysis"},{"key":"ref26","first-page":"436","article-title":"Cryptocurrency volatility forecasting using commonality in intraday volatility","author":"E Djanga","year":"2023","journal-title":"Proceedings of the Fourth ACM International Conference on AI in Finance"},{"issue":"1","key":"ref27","doi-asserted-by":"crossref","first-page":"54","DOI":"10.1093\/jjfinec\/nbr005","article-title":"Forecasting intraday volatility in the U.S. equity market: multiplicative component GARCH","volume":"10","author":"R F Engle","year":"2012","journal-title":"Journal of Financial Econometrics"},{"issue":"2","key":"ref28","doi-asserted-by":"crossref","first-page":"654","DOI":"10.1016\/j.ejor.2017.11.054","article-title":"Deep learning with long short-term memory networks for financial market predictions","volume":"270","author":"T Fischer","year":"2018","journal-title":"European Journal of Operational Research"},{"key":"ref29","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-08332-2","author":"P K Friz","year":"2014","journal-title":"A course on rough paths"},{"key":"ref30","doi-asserted-by":"crossref","DOI":"10.1016\/j.eswa.2023.119879","article-title":"Real-time forecasting of time series in financial markets using sequentially trained dual-LSTMs","volume":"223","author":"K Gajamannage","year":"2023","journal-title":"Expert Systems with Applications"},{"issue":"6","key":"ref31","doi-asserted-by":"crossref","first-page":"933","DOI":"10.1080\/14697688.2017.1393551","volume":"18","author":"J Gatheral","year":"2018","journal-title":"Volatility is rough. Quantitative Finance"},{"key":"ref32","article-title":"Cross-market volatility forecasting with attentionbased spatial-temporal graph convolutional networks","author":"J Gong","year":"2025","journal-title":"Journal of Empirical Finance"},{"issue":"5","key":"ref33","doi-asserted-by":"crossref","first-page":"2223","DOI":"10.1093\/rfs\/hhaa009","article-title":"Empirical asset pricing via machine learning","volume":"33","author":"S Gu","year":"2020","journal-title":"The Review of Financial Studies"},{"issue":"2","key":"ref34","doi-asserted-by":"crossref","first-page":"453","DOI":"10.3982\/ECTA5771","article-title":"The model confidence set","volume":"79","author":"P R Hansen","year":"2011","journal-title":"Econometrica"},{"key":"ref35","doi-asserted-by":"crossref","DOI":"10.1016\/j.ribaf.2021.101402","article-title":"Volatility spillover between economic sectors in financial crisis prediction: evidence spanning the great financial crisis and COVID-19 pandemic","volume":"57","author":"R Laborda","year":"2021","journal-title":"Research in International Business and Finance"},{"issue":"7","key":"ref36","first-page":"6248","article-title":"Automated volatility forecasting","volume":"71","author":"S Z Li","year":"2025","journal-title":"Management Science"},{"issue":"1","key":"ref37","doi-asserted-by":"crossref","first-page":"99","DOI":"10.1080\/1540496X.2023.2210719","article-title":"The time secret of chinese A-share systematic risk: Overnight and intraday","volume":"60","author":"W Liu","year":"2024","journal-title":"Emerging Markets Finance and Trade"},{"key":"ref38","first-page":"81","article-title":"Realized cryptocurrency volatility forecasting with jumps: An empirical study","author":"M.-J Lu","year":"2025","journal-title":"Handbook of Blockchain Analytics"},{"key":"ref39","author":"T Lyons","year":"2014","journal-title":"Rough paths, signatures and the modelling of functions on streams"},{"key":"ref40","doi-asserted-by":"crossref","DOI":"10.1093\/acprof:oso\/9780198506485.001.0001","author":"T Lyons","year":"2002","journal-title":"System control and rough paths"},{"issue":"8","key":"ref41","doi-asserted-by":"crossref","first-page":"1105","DOI":"10.1080\/14697688.2024.2387222","article-title":"Deepvol: volatility forecasting from high-frequency data with dilated causal convolutions","volume":"24","author":"F Moreno-Pino","year":"2024","journal-title":"Quantitative Finance"},{"issue":"3","key":"ref42","doi-asserted-by":"crossref","first-page":"703","DOI":"10.2307\/1913610","article-title":"A simple, positive semi-definite, heteroskedasticity and autocorrelation consistent covariance matrix","volume":"55","author":"W K Newey","year":"1987","journal-title":"Econometrica"},{"key":"ref43","article-title":"Stock market forecasting using a multi-task approach integrating long short-term memory and the random forest framework","volume":"114","author":"H J Park","year":"2022","journal-title":"Applied Soft Computing"},{"issue":"1","key":"ref44","doi-asserted-by":"crossref","first-page":"285","DOI":"10.1007\/s10479-024-06266-0","article-title":"Dynamic asset allocation with asset-specific regime forecasts","volume":"346","author":"Y Shu","year":"2025","journal-title":"Annals of Operations Research"},{"key":"ref45","article-title":"Novel optimization approach for realized volatility forecast of stock price index based on deep reinforcement learning model","volume":"233","author":"Y Yu","year":"2023","journal-title":"Expert Systems with Applications"},{"issue":"2","key":"ref46","doi-asserted-by":"crossref","first-page":"492","DOI":"10.1093\/jjfinec\/nbad005","article-title":"Volatility forecasting with machine learning and intraday commonality","volume":"22","author":"C Zhang","year":"2024","journal-title":"Journal of Financial Econometrics"},{"key":"ref47","doi-asserted-by":"crossref","DOI":"10.1016\/j.eswa.2020.113609","article-title":"A novel deep learning framework: prediction and analysis of financial time series using ceemd and lstm","volume":"159","author":"Y Zhang","year":"2020","journal-title":"Expert Systems with Applications"},{"key":"ref48","doi-asserted-by":"crossref","first-page":"251","DOI":"10.1016\/j.jempfin.2023.07.003","article-title":"Forecasting realized volatility with machine learning: Panel data perspective","volume":"73","author":"H Zhu","year":"2023","journal-title":"Journal of Empirical Finance"},{"key":"ref49","author":"Cluster Cluster"},{"key":"ref50","author":"Cluster Cluster"},{"issue":"7","key":"ref51","doi-asserted-by":"crossref","first-page":"2729","DOI":"10.1093\/rfs\/hhy041","article-title":"Risk everywhere: modeling and managing volatility","volume":"31","author":"T Bollerslev","year":"2018","journal-title":"The Review of Financial Studies"},{"issue":"1","key":"ref52","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.jeconom.2015.10.007","article-title":"Exploiting the errors: A simple approach for improved volatility forecasting","volume":"192","author":"T Bollerslev","year":"2016","journal-title":"Journal of Econometrics"},{"issue":"5","key":"ref53","doi-asserted-by":"crossref","first-page":"2223","DOI":"10.1093\/rfs\/hhaa009","article-title":"Empirical asset pricing via machine learning","volume":"33","author":"S Gu","year":"2020","journal-title":"The Review of Financial Studies"},{"issue":"2","key":"ref54","doi-asserted-by":"crossref","first-page":"453","DOI":"10.3982\/ECTA5771","article-title":"The model confidence set","volume":"79","author":"P R Hansen","year":"2011","journal-title":"Econometrica"},{"issue":"4","key":"ref55","doi-asserted-by":"crossref","first-page":"622","DOI":"10.1016\/j.ijforecast.2018.04.007","article-title":"Forecasting crude oil price volatility","volume":"34","author":"A M Herrera","year":"2018","journal-title":"International Journal of Forecasting"},{"issue":"1","key":"ref56","doi-asserted-by":"crossref","first-page":"246","DOI":"10.1016\/j.jeconom.2010.03.034","article-title":"Volatility forecast comparison using imperfect volatility proxies","volume":"160","author":"A J Patton","year":"2011","journal-title":"Journal of Econometrics"},{"issue":"2","key":"ref57","doi-asserted-by":"crossref","first-page":"492","DOI":"10.1093\/jjfinec\/nbad005","article-title":"Volatility forecasting with machine learning and intraday commonality","volume":"22","author":"C Zhang","year":"2024","journal-title":"Journal of Financial Econometrics"}],"container-title":[],"original-title":[],"deposited":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T13:15:24Z","timestamp":1792761324000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ssrn.com\/abstract=6287198"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":57,"URL":"https:\/\/doi.org\/10.2139\/ssrn.6287198","relation":{},"subject":[],"published":{"date-parts":[[2026]]},"subtype":"preprint"}}