Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/48004
Title: Statistical learning methods for combining technical trading rules and predicting the stock markets
Authors: Fernández Rodríguez, Fernando 
Andrada Félix, Julián 
Acosta González, Eduardo 
UNESCO Clasification: 5302 Econometría
Keywords: Bolsa de valores
Modelos económetricos
Issue Date: 2010
Journal: Data Mining and Management
Abstract: In order to avoid the predictive mismatching that exists between different technical trading rules for predicting stock process, new rules capable of using all the information offered for a wide variety of rules are provided. The system for combining the different types of predictions given by trading rules is based on statistical learning methods (Boosting, and several model averaging methods like Bayesian, or simple averaging methods). When applied to simple and exponential moving average rules designed for predicting the Spanish IBEX35 stock index from January 1998 to December 2007, statistical learning methods supply better out-ofsample results than most of the single moving average rules. In general, the trading strategies based on filtered statistical learning methods are considerably superior to B&H strategy and the behaviour of the filtered statistical learning models reveal a better performance in terms of economic fitness measures during the movements of IBEX35 in clear up-trends and down-trends.
URI: http://hdl.handle.net/10553/48004
ISBN: 9781607412892
Source: Data Mining and Management, p. 141-158
Appears in Collections:Capítulo de libro
Show full item record

Page view(s)

7
checked on Mar 8, 2021

Google ScholarTM

Check

Altmetric


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.