{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T18:45:01Z","timestamp":1762022701114,"version":"build-2065373602"},"reference-count":27,"publisher":"Springer Science and Business Media LLC","issue":"18","license":[{"start":{"date-parts":[[2020,4,20]],"date-time":"2020-04-20T00:00:00Z","timestamp":1587340800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2020,4,20]],"date-time":"2020-04-20T00:00:00Z","timestamp":1587340800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100000266","name":"Engineering and Physical Sciences Research Council","doi-asserted-by":"publisher","award":["1642370"],"award-info":[{"award-number":["1642370"]}],"id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000269","name":"Economic and Social Research Council","doi-asserted-by":"publisher","award":["1535288"],"award-info":[{"award-number":["1535288"]}],"id":[{"id":"10.13039\/501100000269","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Neural Comput &amp; Applic"],"published-print":{"date-parts":[[2020,9]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Much of modern practice in financial forecasting relies on <jats:italic>technicals<\/jats:italic>, an umbrella term for several heuristics applying visual pattern recognition to price charts. Despite its ubiquity in financial media, the reliability of its signals remains a contentious and highly subjective form of \u2018domain knowledge\u2019. We investigate the predictive value of patterns in financial time series, applying machine learning and signal processing techniques to 22 years of US equity data. By reframing technical analysis as a poorly specified, arbitrarily preset feature-extractive layer in a deep neural network, we show that better convolutional filters can be learned directly from the data, and provide visual representations of the features being identified. We find that an ensemble of shallow, thresholded convolutional neural networks optimised over different resolutions achieves state-of-the-art performance on this domain, outperforming technical methods while retaining some of their interpretability.<\/jats:p>","DOI":"10.1007\/s00521-020-04877-9","type":"journal-article","created":{"date-parts":[[2020,4,20]],"date-time":"2020-04-20T09:03:17Z","timestamp":1587373397000},"page":"15249-15262","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":15,"title":["Thresholded ConvNet ensembles: neural networks for technical forecasting"],"prefix":"10.1007","volume":"32","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9907-6130","authenticated-orcid":false,"given":"Sid","family":"Ghoshal","sequence":"first","affiliation":[]},{"given":"Stephen","family":"Roberts","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2020,4,20]]},"reference":[{"issue":"1","key":"4877_CR1","doi-asserted-by":"publisher","first-page":"153","DOI":"10.1111\/j.1540-6261.1994.tb04424.x","volume":"49","author":"L Blume","year":"1994","unstructured":"Blume L, Easley D, O\u2019Hara M (1994) Market statistics and technical analysis: the role of volume. J Financ 49(1):153\u2013181","journal-title":"J Financ"},{"issue":"4","key":"4877_CR2","doi-asserted-by":"publisher","first-page":"405","DOI":"10.2307\/2331231","volume":"32","author":"C Neely","year":"1997","unstructured":"Neely C, Weller P, Dittmar R (1997) Is technical analysis in the foreign exchange market profitable? A genetic programming approach. J Financ Quant Anal 32(4):405\u2013426","journal-title":"J Financ Quant Anal"},{"key":"4877_CR3","doi-asserted-by":"publisher","first-page":"245","DOI":"10.1016\/S0304-405X(98)00052-X","volume":"51","author":"F Allen","year":"1999","unstructured":"Allen F, Karjalainen R (1999) Using genetic algorithms to find technical trading rules. J Financ Econ 51:245\u2013271","journal-title":"J Financ Econ"},{"issue":"2","key":"4877_CR4","doi-asserted-by":"publisher","first-page":"689","DOI":"10.1016\/j.ejor.2016.10.031","volume":"259","author":"C Krauss","year":"2017","unstructured":"Krauss C, Do XA, Huck N (2017) Deep neural networks, gradient-boosted trees, random forests: statistical Arbitrage on the S&P500. Eur J Oper Res 259(2):689\u2013702","journal-title":"Eur J Oper Res"},{"issue":"3\u20134","key":"4877_CR5","first-page":"67","volume":"6","author":"M Dixon","year":"2016","unstructured":"Dixon M, Klabjan D, Bang JH (2016) Classification-based financial markets prediction using deep neural networks. Algorithmic Finance 6(3\u20134):67\u201377","journal-title":"Algorithmic Finance"},{"issue":"5","key":"4877_CR6","first-page":"1706","volume":"55","author":"AW Lo","year":"2000","unstructured":"Lo AW, Mamaysky H, Wang J (2000) Foundations of technical analysis: computational algorithms, statistical inference, and empirical implementation. J Financ 55(5):1706\u20131765","journal-title":"J Financ"},{"issue":"3","key":"4877_CR7","doi-asserted-by":"publisher","first-page":"304","DOI":"10.1016\/0261-5606(92)90048-3","volume":"11","author":"MP Taylor","year":"1992","unstructured":"Taylor MP, Allen H (1992) The use of technical analysis in the foreign exchange market. J Int Money Financ 11(3):304\u2013314","journal-title":"J Int Money Financ"},{"issue":"8","key":"4877_CR8","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","volume":"9","author":"S Hochreiter","year":"1997","unstructured":"Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9(8):1735\u20131780","journal-title":"Neural Comput"},{"key":"4877_CR9","doi-asserted-by":"crossref","unstructured":"Tsantekidis A, Passalis N, Tefas A, Kanniainen J, Gabbouj M, Iosifidis A (2017) Forecasting stock prices from the limit order book using convolutional neural networks. In: 19th IEEE conference on business informatics (CBI) vol 1, pp 7\u201312","DOI":"10.1109\/CBI.2017.23"},{"issue":"2","key":"4877_CR10","doi-asserted-by":"publisher","first-page":"654","DOI":"10.1016\/j.ejor.2017.11.054","volume":"270","author":"T Fischer","year":"2017","unstructured":"Fischer T, Krauss C (2017) Deep learning with long short-term memory networks for financial market predictions. Eur J Oper Res 270(2):654\u2013669","journal-title":"Eur J Oper Res"},{"key":"4877_CR11","doi-asserted-by":"publisher","first-page":"277","DOI":"10.1016\/j.jocs.2017.08.018","volume":"24","author":"M Dixon","year":"2018","unstructured":"Dixon M (2018) Sequence classification of the limit order book using recurrent neural networks. J Comput Sci 24:277\u2013286","journal-title":"J Comput Sci"},{"issue":"11","key":"4877_CR12","doi-asserted-by":"publisher","first-page":"3001","DOI":"10.1109\/TSP.2019.2907260","volume":"67","author":"Z Zhang","year":"2018","unstructured":"Zhang Z, Zohren S, Roberts S (2018) DeepLOB: deep convolutional neural networks for limit order books. IEEE Trans Signal Process 67(11):3001\u20133012","journal-title":"IEEE Trans Signal Process"},{"key":"4877_CR13","doi-asserted-by":"publisher","unstructured":"Sirignano J, Cont R (2018) Universal features of price formation in financial markets: perspectives from deep learning. Available at SSRN: https:\/\/ssrn.com\/abstract=3141294 or https:\/\/doi.org\/10.2139\/ssrn.3141294","DOI":"10.2139\/ssrn.3141294"},{"key":"4877_CR14","first-page":"45","volume":"46","author":"L Romaszko","year":"2015","unstructured":"Romaszko L (2015) Signal correlation prediction using convolutional neural networks. J Mach Learn Res Workshop Conf Proc 46:45\u201356","journal-title":"J Mach Learn Res Workshop Conf Proc"},{"key":"4877_CR15","unstructured":"Krizhevsky A, Sutskever I, Hinton GE (2012) ImageNet classification with deep convolutional neural networks. In: Advances in neural information processing systems, pp 1106\u20131114"},{"issue":"253","key":"4877_CR16","doi-asserted-by":"publisher","first-page":"66","DOI":"10.1080\/01621459.1951.10500769","volume":"46","author":"FJ Massey","year":"1951","unstructured":"Massey FJ (1951) The Kolmogorov\u2013Smirnov test for goodness of fit. J Am Stat Assoc 46(253):66\u201378","journal-title":"J Am Stat Assoc"},{"key":"4877_CR17","unstructured":"Chambers JM, Cleveland WS, Kleiner B, Tukey PA (1983) Comparing data distributions. In: Graphical methods for data analysis, pp 60\u201363"},{"issue":"1","key":"4877_CR18","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1111\/jofi.12365","volume":"71","author":"RD McLean","year":"2016","unstructured":"McLean RD, Pontiff J (2016) Does academic research destroy stock return predictability? J Financ 71(1):5\u201332","journal-title":"J Financ"},{"issue":"6","key":"4877_CR19","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1109\/MSP.2012.2205597","volume":"29","author":"GE Hinton","year":"2012","unstructured":"Hinton GE, Deng L, Yu D, Dahl GE, Mohamed A, Jaitly N, Senior A, Vanhoucke V, Nguyen P, Sainath TN, Kingsbury B (2012) Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups. IEEE Signal Process Mag 29(6):82\u201397","journal-title":"IEEE Signal Process Mag"},{"key":"4877_CR20","unstructured":"Kingma D, Ba J (2015) Adam: a method for stochastic optimization. ICLR 2015"},{"key":"4877_CR21","first-page":"1929","volume":"15","author":"N Srivastava","year":"2014","unstructured":"Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15:1929\u20131958","journal-title":"J Mach Learn Res"},{"key":"4877_CR22","doi-asserted-by":"publisher","first-page":"2145","DOI":"10.1256\/003590002320603584","volume":"128","author":"SJ Mason","year":"2002","unstructured":"Mason SJ, Graham NE (2002) Areas beneath the relative operating characteristics (ROC) and relative operating levels (ROL) curves: statistical significance and interpretation. Q J R Meteorol Soc 128:2145\u20132166","journal-title":"Q J R Meteorol Soc"},{"key":"4877_CR23","first-page":"2342","volume":"37","author":"R Jozefowicz","year":"2015","unstructured":"Jozefowicz R, Zaremba W, Sutskever I (2015) An empirical exploration of recurrent network architectures. J Mach Learn Res 37:2342\u20132350","journal-title":"J Mach Learn Res"},{"key":"4877_CR24","unstructured":"Chung J, Gulcehre C, Cho K, Bengio Y (2014) Empirical evaluation of gated recurrent neural networks on sequence modeling. In: NIPS 2014 deep learning and representation learning workshop"},{"issue":"1","key":"4877_CR25","doi-asserted-by":"publisher","first-page":"307","DOI":"10.1016\/S0925-2312(03)00372-2","volume":"55","author":"K Kim","year":"2003","unstructured":"Kim K (2003) Financial time series forecasting using support vector machines. Neurocomputing 55(1):307\u2013319","journal-title":"Neurocomputing"},{"issue":"20","key":"4877_CR26","doi-asserted-by":"publisher","first-page":"7046","DOI":"10.1016\/j.eswa.2015.05.013","volume":"42","author":"M Ballings","year":"2015","unstructured":"Ballings M, Van den Poel D, Hespeels N, Gryp R (2015) Evaluating multiple classifiers for stock price direction prediction. Exp Syst Appl 42(20):7046\u20137056","journal-title":"Exp Syst Appl"},{"key":"4877_CR27","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1016\/j.patrec.2015.11.011","volume":"70","author":"G Antipova","year":"2016","unstructured":"Antipova G, Berrani SA, Dugelay JL (2016) Minimalistic CNN-based ensemble model for gender prediction from face images. Pattern Recognit Lett 70:59\u201365","journal-title":"Pattern Recognit Lett"}],"container-title":["Neural Computing and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00521-020-04877-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00521-020-04877-9\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00521-020-04877-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,4,20]],"date-time":"2021-04-20T00:03:12Z","timestamp":1618876992000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00521-020-04877-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,4,20]]},"references-count":27,"journal-issue":{"issue":"18","published-print":{"date-parts":[[2020,9]]}},"alternative-id":["4877"],"URL":"https:\/\/doi.org\/10.1007\/s00521-020-04877-9","relation":{},"ISSN":["0941-0643","1433-3058"],"issn-type":[{"type":"print","value":"0941-0643"},{"type":"electronic","value":"1433-3058"}],"subject":[],"published":{"date-parts":[[2020,4,20]]},"assertion":[{"value":"22 February 2019","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"20 March 2020","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"20 April 2020","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Compliance with ethical standards"}},{"value":"The authors declare that they have no conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}]}}