Skip navigation
Por favor, use este identificador para citar o enlazar este ítem: http://repositorio.unb.br/handle/10482/45128
Ficheros en este ítem:
No hay ficheros asociados a este ítem.
Título : Reaction trend system with GARCH quantiles as action points
Autor : Fiorucci, Jose Augusto
Silva, Geraldo Nunes
Barboza, Flavio Luiz de Moraes
metadata.dc.contributor.email: mailto:jafiorucci@unb.br
mailto:geraldo.silva@unesp.br
mailto:flmbarboza@ufu.br
metadata.dc.identifier.orcid: https://orcid.org/0000-0002-1201-9089
https://orcid.org/0000-0002-3574-9893
https://orcid.org/0000-0002-3449-5297
Assunto:: Modelo estatístico
Volatilidade
Intervalo de preços
Modelos ARCH
Fecha de publicación : 5-mar-2022
Editorial : Elsevier
Citación : FIORUCCI, Jose Augusto; SILVA, Geraldo Nunes; BARBOZA, Flavio. Reaction trend system with GARCH quantiles as action points. Expert Systems with Applications, v. 198, art. 116750, jul. 2022. DOI 10.1016/j.eswa.2022.116750. Disponível em: https://www.sciencedirect.com/science/article/pii/S0957417422002172?via%3Dihub. Acesso em: 07 nov. 2022.
Abstract: Most trading systems developed from technical indicators are designed to operate in either trending or non-trending markets but they are rarely useful for both markets. A reaction trend system allows operations in both markets. In this study, we improve the calculus of four action points using a statistical volatility model and then replace the action points with those derived from GARCH quantiles. Although the proposed system is more advanced, it maintains the same operational logic as the method proposed by Wilder. Empirical tests on various assets suggest that the novel method performs better. Finally, we compare the performance of our proposed system with previous studies that analysed assets in different markets and regions. Our system has been proven consistently competitive in all situations.
DOI: https://doi.org/10.1016/j.eswa.2022.116750
metadata.dc.relation.publisherversion: https://www.sciencedirect.com/science/article/pii/S0957417422002172?via%3Dihub
Aparece en las colecciones: Artigos publicados em periódicos e afins

Mostrar el registro Dublin Core completo del ítem " class="statisticsLink btn btn-primary" href="/jspui/handle/10482/45128/statistics">



Los ítems de DSpace están protegidos por copyright, con todos los derechos reservados, a menos que se indique lo contrario.