{"entities":{"Q4806033":{"pageid":6835613,"ns":120,"title":"Item:Q4806033","lastrevid":109045709,"modified":"2026-11-11T07:25:02Z","type":"item","id":"Q4806033","labels":{"en":{"language":"en","value":"Neural networks and financial markets. Predicting, combining and portfolio optimisation"}},"descriptions":{"en":{"language":"en","value":"scientific article; zbMATH DE number 1903876"}},"aliases":{},"claims":{"P31":[{"mainsnak":{"snaktype":"value","property":"P31","hash":"fd5912e4dab4b881a8eb0eb27e7893fef55176ad","datavalue":{"value":{"entity-type":"item","numeric-id":56887,"id":"Q56887"},"type":"wikibase-entityid"},"datatype":"wikibase-item"},"type":"statement","id":"Q4806033$32FA9775-2B2A-4628-8459-B09C194F54AE","rank":"normal"}],"P225":[{"mainsnak":{"snaktype":"value","property":"P225","hash":"bd8653d1fa810fe2c1c8536ee964567b3d449589","datavalue":{"value":"1022.91031","type":"string"},"datatype":"external-id"},"type":"statement","id":"Q4806033$8F0E873A-2AFA-4A7A-B8CE-C47E315D2E27","rank":"normal"}],"P28":[{"mainsnak":{"snaktype":"value","property":"P28","hash":"c9741052acb3601490f70dda00f31b7265d37a65","datavalue":{"value":{"time":"+2003-05-04T00:00:00Z","timezone":0,"before":0,"after":0,"precision":11,"calendarmodel":"http://www.wikidata.org/entity/Q1985727"},"type":"time"},"datatype":"time"},"type":"statement","id":"Q4806033$215FB9FD-7BE1-4366-8E49-BAEBD6271E73","rank":"normal"}],"P226":[{"mainsnak":{"snaktype":"value","property":"P226","hash":"7f918b855cc4e1ef3c50dee94916276dca231342","datavalue":{"value":"91G10","type":"string"},"datatype":"external-id"},"type":"statement","id":"Q4806033$E2BCC7DD-A983-4F0F-A88C-A7CB93AAD6EF","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P226","hash":"cfe779e91fe9c53ee133568259955801965765ae","datavalue":{"value":"68T05","type":"string"},"datatype":"external-id"},"type":"statement","id":"Q4806033$AC0EFBA3-41A9-48CA-B46C-04577BF17C38","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P226","hash":"6e419163fd9b5f210c8c0c28cdb07c2197a2bcac","datavalue":{"value":"91B26","type":"string"},"datatype":"external-id"},"type":"statement","id":"Q4806033$29DFEB68-FD90-44B5-A7B1-49EC54E1F208","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P226","hash":"325e035d1812352c3a83010cdc35176514b19a00","datavalue":{"value":"91B84","type":"string"},"datatype":"external-id"},"type":"statement","id":"Q4806033$D2FAF71A-40D8-4225-87EF-D1AF2982BF16","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P226","hash":"5ac4eec6d7cc8329ab3df374138b13cb9d331467","datavalue":{"value":"82B20","type":"string"},"datatype":"external-id"},"type":"statement","id":"Q4806033$7B987881-5A9A-42AB-BF7F-3DC9B3725930","rank":"normal"}],"P1451":[{"mainsnak":{"snaktype":"value","property":"P1451","hash":"2d4b9be2af19babf6dd87daf6ab8cc8f7ce319d6","datavalue":{"value":"1903876","type":"string"},"datatype":"external-id"},"type":"statement","id":"Q4806033$00C5D29D-60DC-47EB-AC3B-674B827E20B3","rank":"normal"}],"P1450":[{"mainsnak":{"snaktype":"value","property":"P1450","hash":"c8d0c667a320ca57c4da94e0b00c5e5c35afae41","datavalue":{"value":"Neural networks","type":"string"},"datatype":"string"},"type":"statement","id":"Q4806033$59E56EBC-7613-46C4-A8EB-E197F6CD6357","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1450","hash":"25209a4c99aaf5de18cde0a06b46919a853b5556","datavalue":{"value":"Financial markets","type":"string"},"datatype":"string"},"type":"statement","id":"Q4806033$63664260-C79B-407F-9CA8-1CC38F75D26D","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1450","hash":"df48535ca9cf5dde523ae75db813902dbe313678","datavalue":{"value":"Portfolio optimisation","type":"string"},"datatype":"string"},"type":"statement","id":"Q4806033$54F89CE9-1F2E-4B76-951D-3E4634EF9E23","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1450","hash":"35f0adb7aaad8ebf302b3890169a0f7eb033c179","datavalue":{"value":"predicting financial market returns","type":"string"},"datatype":"string"},"type":"statement","id":"Q4806033$698D70E4-01BB-4660-A424-BC7DF51F67BC","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1450","hash":"8fc672ea664500ee0c6b559b35536eaa5db4aa14","datavalue":{"value":"bond yield curve","type":"string"},"datatype":"string"},"type":"statement","id":"Q4806033$4879FA6D-C885-4310-A7D8-647BDDF3B965","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1450","hash":"d73dfddd0f425a8e08f008b7cb0f9841f3c99e33","datavalue":{"value":"asset value prediction","type":"string"},"datatype":"string"},"type":"statement","id":"Q4806033$20E64532-1D97-4225-903A-585CE0A5D4A9","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1450","hash":"e5a53c294ba64bd438aa5380f9fa07c165a64b19","datavalue":{"value":"neural network prediction models","type":"string"},"datatype":"string"},"type":"statement","id":"Q4806033$A43D862B-E525-4636-A6B2-3DA5E7CF6ADA","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1450","hash":"6ff7df13e38c40e0baf50ad39a206a3998bfd605","datavalue":{"value":"Bayesian evidence ratio model","type":"string"},"datatype":"string"},"type":"statement","id":"Q4806033$C5779EB3-74A5-4743-8159-E3129D731DEC","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1450","hash":"b31c38f57efa1884ba435f057eef06d7caeb0e1e","datavalue":{"value":"interest series prediction","type":"string"},"datatype":"string"},"type":"statement","id":"Q4806033$3BEDCE0A-4314-4ECD-ACA8-E88AE36D1986","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1450","hash":"be6d6509af6e36cc60b3a2e33bf829282114df47","datavalue":{"value":"bond series prediction","type":"string"},"datatype":"string"},"type":"statement","id":"Q4806033$A6D73E29-5CCD-4583-9811-BA16BB835187","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1450","hash":"98f35b57f2a360658216248a468708bf3edb3f38","datavalue":{"value":"portfolio allocation and optimization","type":"string"},"datatype":"string"},"type":"statement","id":"Q4806033$5152CFEA-5052-4D8B-B161-2D281338C379","rank":"normal"}],"P1460":[{"mainsnak":{"snaktype":"value","property":"P1460","hash":"57f7fea50d2ce1b39b695c4a1313582eed405e38","datavalue":{"value":{"entity-type":"item","numeric-id":5976449,"id":"Q5976449"},"type":"wikibase-entityid"},"datatype":"wikibase-item"},"type":"statement","id":"Q4806033$F55B6CC9-9078-4E77-B458-C09D47DFEDC4","rank":"normal"}],"P1643":[{"mainsnak":{"snaktype":"value","property":"P1643","hash":"57dcf9e0838939190d4e0bcb192833a2ac5e5d56","datavalue":{"value":{"entity-type":"item","numeric-id":4994396,"id":"Q4994396"},"type":"wikibase-entityid"},"datatype":"wikibase-item"},"type":"statement","qualifiers":{"P1659":[{"snaktype":"value","property":"P1659","hash":"3acf498668037ed19eea6be3c2caf4dba22396b4","datavalue":{"value":{"amount":"+0.7905347347259521","unit":"1"},"type":"quantity"},"datatype":"quantity"}],"P1660":[{"snaktype":"value","property":"P1660","hash":"a327a09ea0305e98d5cf33bd4036320e19f2aed0","datavalue":{"value":{"entity-type":"item","numeric-id":6821328,"id":"Q6821328"},"type":"wikibase-entityid"},"datatype":"wikibase-item"}]},"qualifiers-order":["P1659","P1660"],"id":"Q4806033$01B232A3-1FD5-484A-BD7A-85DF8FDC2C2C","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1643","hash":"63e7ed789ef9d7b9fb36855a01727db9f3ac9982","datavalue":{"value":{"entity-type":"item","numeric-id":4351325,"id":"Q4351325"},"type":"wikibase-entityid"},"datatype":"wikibase-item"},"type":"statement","qualifiers":{"P1659":[{"snaktype":"value","property":"P1659","hash":"cc95caf538c843c606dee35368d4dcc5f0cb87f8","datavalue":{"value":{"amount":"+0.7713019847869873","unit":"1"},"type":"quantity"},"datatype":"quantity"}],"P1660":[{"snaktype":"value","property":"P1660","hash":"a327a09ea0305e98d5cf33bd4036320e19f2aed0","datavalue":{"value":{"entity-type":"item","numeric-id":6821328,"id":"Q6821328"},"type":"wikibase-entityid"},"datatype":"wikibase-item"}]},"qualifiers-order":["P1659","P1660"],"id":"Q4806033$C906A413-3CB7-4616-9FDD-2F0B4AE9460F","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1643","hash":"fa4de1e4cfcdd170cf7651ff0bffee45742618b5","datavalue":{"value":{"entity-type":"item","numeric-id":1683384,"id":"Q1683384"},"type":"wikibase-entityid"},"datatype":"wikibase-item"},"type":"statement","qualifiers":{"P1659":[{"snaktype":"value","property":"P1659","hash":"51717f2e9ec4993ce2a90ab5221e57a3219bf212","datavalue":{"value":{"amount":"+0.7712475657463074","unit":"1"},"type":"quantity"},"datatype":"quantity"}],"P1660":[{"snaktype":"value","property":"P1660","hash":"a327a09ea0305e98d5cf33bd4036320e19f2aed0","datavalue":{"value":{"entity-type":"item","numeric-id":6821328,"id":"Q6821328"},"type":"wikibase-entityid"},"datatype":"wikibase-item"}]},"qualifiers-order":["P1659","P1660"],"id":"Q4806033$2432D5CC-03E1-4C0F-883E-D1201B2ED116","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1643","hash":"a917bffafe4aaa4c3a79514f1a2967f4bf4821d3","datavalue":{"value":{"entity-type":"item","numeric-id":3376484,"id":"Q3376484"},"type":"wikibase-entityid"},"datatype":"wikibase-item"},"type":"statement","qualifiers":{"P1659":[{"snaktype":"value","property":"P1659","hash":"4cbba6333c85937a9215bb551989dc70991aebcc","datavalue":{"value":{"amount":"+0.7655978798866272","unit":"1"},"type":"quantity"},"datatype":"quantity"}],"P1660":[{"snaktype":"value","property":"P1660","hash":"a327a09ea0305e98d5cf33bd4036320e19f2aed0","datavalue":{"value":{"entity-type":"item","numeric-id":6821328,"id":"Q6821328"},"type":"wikibase-entityid"},"datatype":"wikibase-item"}]},"qualifiers-order":["P1659","P1660"],"id":"Q4806033$2E5BCC0E-0818-4901-A570-BB2FD8E6C71F","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P1643","hash":"620737bf2831527582d1f83ac2ac5a35457e7801","datavalue":{"value":{"entity-type":"item","numeric-id":4358557,"id":"Q4358557"},"type":"wikibase-entityid"},"datatype":"wikibase-item"},"type":"statement","qualifiers":{"P1659":[{"snaktype":"value","property":"P1659","hash":"784136bda857570b23d859f485a51dee238585a2","datavalue":{"value":{"amount":"+0.7615711092948914","unit":"1"},"type":"quantity"},"datatype":"quantity"}],"P1660":[{"snaktype":"value","property":"P1660","hash":"a327a09ea0305e98d5cf33bd4036320e19f2aed0","datavalue":{"value":{"entity-type":"item","numeric-id":6821328,"id":"Q6821328"},"type":"wikibase-entityid"},"datatype":"wikibase-item"}]},"qualifiers-order":["P1659","P1660"],"id":"Q4806033$5882BD41-12AE-4ECE-8788-F4EFE5427839","rank":"normal"}],"P159":[{"mainsnak":{"snaktype":"value","property":"P159","hash":"fe68c5ce83bc418bc2daab5e4e874eec673dc684","datavalue":{"value":{"text":"Neural networks and financial markets. Predicting, combining and portfolio optimisation","language":"en"},"type":"monolingualtext"},"datatype":"monolingualtext"},"type":"statement","id":"Q4806033$E05FE553-8D70-4CFD-9E98-96BE1D67BF0A","rank":"normal"}],"P1448":[{"mainsnak":{"snaktype":"value","property":"P1448","hash":"60776269d78f096ff1eb8eee888c314903f73fe1","datavalue":{"value":"This book presents the results of the NewQuant research team (Econostat New Quant Ltd), over a period of 10 years, for predicting financial market returns. The overall process used by the NewQuant team contains components needed in any approach on forecasting of assets. The following stages have been developed by the contributors to this book in order to obtain a final proposal of asset allocations in a portfolio investing either solely in G7 bonds (and cash) or in both G7 bonds and equity indices: data analysis, data selection, model building, prediction combining, portfolio allocation, and advice. The main purpose of the book is to point out new ideas and technologies arising from Artificial Intelligence, to be applied within the mentioned stages in order to improve the financial forecasting process. The book is organized into five parts, grouping 27 chapters. Here it is a brief description of their contents: Part I (Introduction to Prediction in the Financial Markets) contains five chapters that give a broad account for the financial markets. Chapter 1 shows what assets are to be considered, which are the factors that may influence them, and how financial and economic indicators may be used to predict future values of these various assets. Chapter 2 gives the basic structures of the two fundamentally different approaches to prediction: that of pattern matching (univariate), and of indicator driving of assets (multivariate). Chapter 3 reviews the question of market efficiency, with the conclusion that predictability is present as a necessary component of the financial markets. Chapter 4 discusses about bonds, the dependence on maturity of the bond yield (curve) in terms of the time value of money (setting up the structure of the price of a given maturity bond). Chapter 5 deals with input variable selection in constructing multivariate models for the assets of interest. Part II (Theory of Prediction Modeling) encloses chapters 6-10. Chapter 6 surveys the prediction models and outlines the hard work needed to discover a good prediction formula. The prediction model is let to depend on a set of free parameters, chosen to some criterion error itself leading to a quantization problem in parameter choice. Chapter 7 discusses techniques to avoid various dangers in this optimization process (such as overfitting past data). Chapter 8 considers extending the effective data set under the heading of the bootstrap, using bagging and ensembles. Chapter 9 takes into account the linear models as providing a springboard to the langer range of models, including nonlinear ones. Chapter 10 examines selection methods to reduce the set of inputs for detailed model building, and the effectiveness of these methods. Part III (Theory of Specific Prediction Models) is made up of chapters 11-15. Chapter 11 contains an introduction to neural networks as the core modeling technologies for prediction models. Chapter 12 develops learning the trading stategies by reinforcements, using reward/penalty derived from profitability. Chapter 13 analyzes the underlying dynamics of the markets, embedding and dynamical dimensions being extracted from financial time series. Chapter 14 proposes the modeling approach of vector machines, mapping the input vector into a larger dimension space and describing the support vector and relevance vector machines. Chapter 15 elaborates the assessment of the effectiveness of models, using a Bayesian evidence ratio model. Part IV (Prediction Model Applications) exposes in chapters 16-24 the application to explicit prediction of bond and equity series, relying on various modeling technologies. Chapter 16 explains the shape of the bond yield curve, its possible parameterizations, and prediction of the parameter time series. In Chapter 17, for bond prediction is used the relevance vector machine, a linear regression method selecting the inputs. Prediction of the Australian bonds is considered. After using the relevance vector-based method for input selection, Chapter 18 develops feedforward and recurrent neural networks to predict US Government bonds. Results across a range of other markets are discussed. Chapter 19 shows the approach of a specialized neural network, viz. the adaptive lag method with a training rule developed for the best lags. The results are applied to predict German government bond returns. Chapter 20 considers a further neural network, which uses random generation and subsequent selection of neural networks. The resultant weak classifier is applied to US bood returns. Chapter 21 looks at the method of co-integrated time series that allow prediction of the series of interest. This is applied to effective models of prediction for suitable combinations of FTSE 100 stocks. Chapter 22 utilizes the reinforcement training technique of Chapter 12 to a portfolio of assets. Chapter 23 contains results obtained from various univariate models. The considered models, i.e. Nearest Neighbours, Group Method of Data Handling (GMDH), Support Vector Machine and Relevance Vector Machine, are used to predict various bond series (UK, European, German). Chapter 24 develops the combining of predictions across a number of model technologies. The final Part V (Optimizing and Beyond) indicates in chapters 25-27 how are made the allocations to given markets across a given portfolio. In Chapter 25 is elaborated a quadratic optimizer for the allocation process, while Chapter 26 describes nonlinear modeling techniques such as neural networks used to model dynamics of trading, as well as various strategies and predictions made by investors. The final Chapter 27 covers a summary of the book, its approach philosophy, the associated questions raised, and the possible future research directions. This remarkable book is addressed equally to researchers, quantitative fund managers, and any other professionals involved in financial prediction, time series analysis and information processing.","type":"string"},"datatype":"string"},"type":"statement","id":"Q4806033$F7C87E96-F026-49B0-AAFC-29D7D72B02C6","rank":"normal"}]},"sitelinks":{"mardi":{"site":"mardi","title":"Neural networks and financial markets. Predicting, combining and portfolio optimisation","badges":[],"url":"https://portal.mardi4nfdi.de/wiki/Neural_networks_and_financial_markets._Predicting,_combining_and_portfolio_optimisation"}}}}}