{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,20]],"date-time":"2026-10-20T10:39:26Z","timestamp":1792492766805,"version":"4.5.0"},"posted":{"date-parts":[[2026]]},"group-title":"SSRN","reference-count":43,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/www.uspto.gov\/ip-policy\/copyright-policy\/copyright-basics"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>This study examines the predictive performance of Transformer-based stochastic discount factor models over one-, three-, six-, and twelve-month horizons. By sharing information across assets, the self-attention mechanism captures conditional pricing signals that conventional machine learning approaches systematically overlook. When benchmarked against a multilayer perceptron, the Transformer achieves higher out-of-sample Sharpe ratios at every horizon over the full out-of-sample period, both gross and net of transaction costs, and sustains substantially higher break-even trading costs. For both models, performance exhibits a monotonically declining trend with the forecast horizon. Our results demonstrate that the economic value of attention extends across the entire term structure of investment horizons.<\/jats:p>","DOI":"10.2139\/ssrn.7605161","type":"posted-content","created":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T10:29:26Z","timestamp":1791973766000},"source":"Crossref","is-referenced-by-count":0,"title":["Multi-Horizon Stock Return Prediction via Transformer-Based Deep Learning"],"prefix":"10.2139","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-5793-9886","authenticated-orcid":true,"given":"Tobias","family":"Cramer","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5247-0405","authenticated-orcid":true,"given":"Wolfgang","family":"Drobetz","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tizian","family":"Otto","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"ref1","first-page":"2587","article-title":"Economic Restrictions: Evidence from Stock Return Predictability","volume":"69","author":"D Avramov","year":"2023","journal-title":"Machine Learning vs"},{"issue":"4","key":"ref2","doi-asserted-by":"crossref","first-page":"40","DOI":"10.3905\/jfds.2023.1.135","article-title":"The Term Structure of Machine Learning Alpha","volume":"5","author":"D Blitz","year":"2023","journal-title":"The Journal of Financial Data Science"},{"key":"ref3","doi-asserted-by":"crossref","first-page":"3411","DOI":"10.1093\/rfs\/hhp003","article-title":"Parametric Portfolio Policies: Exploiting Characteristics in the Cross-Section of Equity Returns","volume":"22","author":"M W Brandt","year":"2009","journal-title":"The Review of Financial Studies"},{"issue":"5","key":"ref4","doi-asserted-by":"crossref","first-page":"2447","DOI":"10.1111\/jofi.13477","article-title":"Forest through the Trees: Building Cross-Sections of Stock Returns","volume":"80","author":"S Bryzgalova","year":"2025","journal-title":"The Journal of Finance"},{"issue":"4","key":"ref5","doi-asserted-by":"crossref","first-page":"119","DOI":"10.3905\/jfds.2023.1.139","article-title":"Predicting Returns with Machine Learning across Horizons, Firm Size, and Time","volume":"5","author":"N Cakici","year":"2023","journal-title":"The Journal of Financial Data Science"},{"issue":"3","key":"ref6","doi-asserted-by":"crossref","first-page":"968","DOI":"10.1017\/S0022109022000874","article-title":"Zeroing In on the Expected Returns of Anomalies","volume":"58","author":"A Y Chen","year":"2023","journal-title":"Journal of Financial and Quantitative Analysis"},{"issue":"2","key":"ref7","doi-asserted-by":"crossref","first-page":"207","DOI":"10.1561\/104.00000112","article-title":"Open Source Cross-Sectional Asset Pricing","volume":"11","author":"A Y Chen","year":"2022","journal-title":"Critical Finance Review"},{"issue":"2","key":"ref8","doi-asserted-by":"crossref","first-page":"714","DOI":"10.1287\/mnsc.2023.4695","article-title":"Deep Learning in Asset Pricing","volume":"70","author":"L Chen","year":"2024","journal-title":"Management Science"},{"issue":"1","key":"ref9","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1016\/j.jacceco.2014.06.001","article-title":"Have Capital Market Anomalies Attenuated in the Recent Era of High Liquidity and Trading Activity?","volume":"58","author":"T Chordia","year":"2014","journal-title":"Journal of Accounting and Economics"},{"issue":"4","key":"ref10","doi-asserted-by":"crossref","first-page":"1977","DOI":"10.1111\/j.1540-6261.2008.01379.x","article-title":"Economic Links and Predictable Returns","volume":"63","author":"L Cohen","year":"2008","journal-title":"The Journal of Finance"},{"issue":"3","key":"ref11","article-title":"Estimating Industry Betas via Machine Learning: Promises and Pitfalls of Multi-Output Predictions","volume":"24","author":"T Cramer","year":"2026","journal-title":"Journal of Investment Management"},{"key":"ref12","doi-asserted-by":"crossref","DOI":"10.3386\/w33012","volume":"33012","author":"A Didisheim","year":"2024","journal-title":"APT or \"AIPT\"? The Surprising Dominance of Large Factor Models. NBER Working Papers"},{"key":"ref13","first-page":"2793","article-title":"Attention is Not All You Need: Pure Attention Loses Rank Doubly Exponentially with Depth","author":"Y Dong","year":"2021","journal-title":"International Conference on Machine Learning"},{"issue":"7","key":"ref14","doi-asserted-by":"crossref","first-page":"507","DOI":"10.1057\/s41260-021-00237-x","article-title":"Empirical Asset Pricing via Machine Learning: Evidence from the European Stock Market","volume":"22","author":"W Drobetz","year":"2021","journal-title":"Journal of Asset Management"},{"issue":"1","key":"ref15","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.jfineco.2014.10.010","article-title":"A Five-Factor Asset Pricing Model","volume":"116","author":"E F Fama","year":"2015","journal-title":"Journal of Financial Economics"},{"issue":"5","key":"ref16","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"},{"key":"ref17","doi-asserted-by":"crossref","first-page":"557","DOI":"10.1111\/j.1540-6261.1997.tb04813.x","article-title":"Assessing Specification Errors in Stochastic Discount Factor Models","volume":"52","author":"L P Hansen","year":"1997","journal-title":"The Journal of Finance"},{"issue":"3","key":"ref18","doi-asserted-by":"crossref","first-page":"587","DOI":"10.2307\/1913601","article-title":"The Role of Conditioning Information in Deducing Testable Restrictions Implied by Dynamic Asset Pricing Models","volume":"55","author":"L P Hansen","year":"1987","journal-title":"Econometrica"},{"issue":"4","key":"ref19","doi-asserted-by":"crossref","first-page":"1113","DOI":"10.1093\/revfin\/hhm003","article-title":"Industry Information Diffusion and the Lead-lag Effect in Stock Returns","volume":"20","author":"K Hou","year":"2007","journal-title":"The Review of Financial Studies"},{"issue":"1","key":"ref20","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1111\/j.1540-6261.1993.tb04702.x","article-title":"Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency","volume":"48","author":"N Jegadeesh","year":"1993","journal-title":"The Journal of Finance"},{"key":"ref21","doi-asserted-by":"crossref","first-page":"2465","DOI":"10.1111\/jofi.13249","article-title":"Is there a Replication Crisis in Finance?","volume":"78","author":"T I Jensen","year":"2023","journal-title":"The Journal of Finance"},{"key":"ref22","first-page":"1","volume":"0","author":"T I Jensen","year":"2026","journal-title":"Machine Learning and the Implementable Efficient Frontier"},{"key":"ref23","first-page":"205","volume":"13","author":"B Kelly","year":"2023","journal-title":"Financial Machine Learning"},{"key":"ref24","doi-asserted-by":"crossref","DOI":"10.3386\/w33351","volume":"33351","author":"B Kelly","year":"2025","journal-title":"Artificial Intelligence Asset Pricing Models. NBER Working Papers"},{"key":"ref25","doi-asserted-by":"crossref","first-page":"459","DOI":"10.1111\/jofi.13298","article-title":"The Virtue of Complexity in Return Prediction","volume":"79","author":"B Kelly","year":"2024","journal-title":"The Journal of Finance"},{"key":"ref26","author":"B Kelly","year":"2019","journal-title":"Instrumented Principal Component Analysis"},{"issue":"2","key":"ref27","doi-asserted-by":"crossref","first-page":"271","DOI":"10.1016\/j.jfineco.2019.06.008","article-title":"Shrinking the Cross-Section","volume":"135","author":"S Kozak","year":"2020","journal-title":"Journal of Financial Economics"},{"issue":"2","key":"ref28","doi-asserted-by":"crossref","first-page":"64","DOI":"10.1016\/j.jfineco.2021.08.017","article-title":"Machine Learning in the Chinese Stock Market","volume":"145","author":"M Leippold","year":"2022","journal-title":"Journal of Financial Economics"},{"issue":"2","key":"ref29","doi-asserted-by":"crossref","first-page":"21","DOI":"10.3905\/jfds.2021.1.062","article-title":"The Promises and Pitfalls of Machine Learning for Predicting Cross-Sectional Stock Returns","volume":"3","author":"E Leung","year":"2021","journal-title":"The Journal of Financial Data Science"},{"key":"ref30","doi-asserted-by":"crossref","first-page":"1748","DOI":"10.1016\/j.ijforecast.2021.03.012","article-title":"Temporal Fusion Transformers for Interpretable Multi-Horizon Time Series Forecasting","volume":"37","author":"B Lim","year":"2021","journal-title":"International Journal of Forecasting"},{"issue":"2","key":"ref31","doi-asserted-by":"crossref","first-page":"175","DOI":"10.1093\/rfs\/3.2.175","article-title":"When are Contrarian Profits Due to Stock Market Overreaction?","volume":"3","author":"A W Lo","year":"1990","journal-title":"The Review of Financial Studies"},{"issue":"1","key":"ref32","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1111\/jofi.12365","article-title":"Does Academic Research Destroy Stock Return Predictability?","volume":"71","author":"R D Mclean","year":"2016","journal-title":"The Journal of Finance"},{"issue":"4","key":"ref33","doi-asserted-by":"crossref","first-page":"1249","DOI":"10.1111\/0022-1082.00146","article-title":"Do Industries Explain Momentum?","volume":"54","author":"T J Moskowitz","year":"1999","journal-title":"The Journal of Finance"},{"issue":"3","key":"ref34","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"},{"issue":"1","key":"ref35","doi-asserted-by":"crossref","first-page":"104","DOI":"10.1093\/rfs\/hhv063","article-title":"A Taxonomy of Anomalies and Their Trading Costs","volume":"29","author":"R Novy-Marx","year":"2016","journal-title":"The Review of Financial Studies"},{"issue":"1","key":"ref36","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1080\/0015198X.2018.1547057","article-title":"Comparing Cost-Mitigation Techniques","volume":"75","author":"R Novy-Marx","year":"2019","journal-title":"Financial Analysts Journal"},{"issue":"2","key":"ref37","doi-asserted-by":"crossref","first-page":"515","DOI":"10.1016\/j.jfineco.2020.02.012","article-title":"What You See is Not What You Get: The Costs of Trading Market Anomalies","volume":"137","author":"A J Patton","year":"2020","journal-title":"Journal of Financial Economics"},{"key":"ref38","article-title":"Attention Is All You Need","volume":"30","author":"A Vaswani","year":"2017","journal-title":"Advances in Neural Information Processing Systems (NIPS)"},{"key":"ref39","doi-asserted-by":"crossref","DOI":"10.3386\/w34861","author":"Y Wang","year":"2026","journal-title":"Machine Learning Meets Markowitz"},{"key":"ref40","article-title":"Machine Learning for Stock Return Prediction: Transformers or Simple Neural Networks","author":"Z Wang","year":"2025","journal-title":"Finance Research Letters"},{"key":"ref41","author":"Q Wen","year":"2022","journal-title":"Transformers in Time Series: A Survey. Proceedings of the 32nd International Joint Conference on Artificial Intelligence (IJCAI 2023)"},{"key":"ref42","first-page":"22419","article-title":"Autoformer: Decomposition Transformers with Auto-Correlation for Long","author":"H Wu","year":"2021","journal-title":"Term Series Forecasting. NIPS'21: Proceedings of the 35th International Conference on Neural Information Processing Systems"},{"issue":"C","key":"ref43","article-title":"Where Have the Profits Gone? Market Efficiency and the Disappearing Equity Anomalies in Country and Industry Returns","author":"A Zaremba","year":"2020","journal-title":"Journal of Banking & Finance"}],"container-title":[],"original-title":[],"deposited":{"date-parts":[[2026,10,20]],"date-time":"2026-10-20T09:50:11Z","timestamp":1792489811000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ssrn.com\/abstract=7605161"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":43,"URL":"https:\/\/doi.org\/10.2139\/ssrn.7605161","relation":{},"subject":[],"published":{"date-parts":[[2026]]},"subtype":"preprint"}}