Tactical asset allocation : a novel approach via machine learning returns prediction
Autor(a) principal: | |
---|---|
Data de Publicação: | 2023 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Texto Completo: | http://hdl.handle.net/10400.14/41426 |
Resumo: | The thesis tries to develop a novel quantitative approach for the development of trading strate gies managed through Tactical Asset Allocation investment style. The work employs a returns prediction performed via supervised learning algorithms and studies whether they can help in the deliverance of superior returns compared with the classical buy-and-hold strategies. The models manage to predict the movement of the markets with an acceptable precision level and try to give hints for bet sizing, playing a key role in the weight's definitions. While most of the previously developed Tactical Asset Allocation strategies were performed through a historical analysis. The novelty of the approach comes from the incorporation of these forward-looking predicted values. In thisstudy, Multi-Layer Perceptron Neural Networks, Support Vector Machines, and Random Forests were used to predict the returns of two main market indexes, respectively S&P500 and Eurostoxx600. The models are used in their regressor form to get a continuous output, expected to be the true value of the returns for the following day. Based on this prediction, several trading strategies have been developed and tested. Results indicate that the proposed approach can give positive signals for what concerns return achievement and reward-to-risk ratio improvement. Nevertheless, due to the high dynamicity of the strategy, as implied by Tactical Asset Allocation hypothesis, transaction costs play a key role in final returns deliverance. All the trading strategies are performed considering the different outcomes of the models. |
id |
RCAP_35914f948893e79db2f3a18caa4a1568 |
---|---|
oai_identifier_str |
oai:repositorio.ucp.pt:10400.14/41426 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Tactical asset allocation : a novel approach via machine learning returns predictionTactical asset allocationMachine learningReturns predictionAlocação tactica de activosPrevisão dos retornosDomínio/Área Científica::Ciências Sociais::Economia e GestãoThe thesis tries to develop a novel quantitative approach for the development of trading strate gies managed through Tactical Asset Allocation investment style. The work employs a returns prediction performed via supervised learning algorithms and studies whether they can help in the deliverance of superior returns compared with the classical buy-and-hold strategies. The models manage to predict the movement of the markets with an acceptable precision level and try to give hints for bet sizing, playing a key role in the weight's definitions. While most of the previously developed Tactical Asset Allocation strategies were performed through a historical analysis. The novelty of the approach comes from the incorporation of these forward-looking predicted values. In thisstudy, Multi-Layer Perceptron Neural Networks, Support Vector Machines, and Random Forests were used to predict the returns of two main market indexes, respectively S&P500 and Eurostoxx600. The models are used in their regressor form to get a continuous output, expected to be the true value of the returns for the following day. Based on this prediction, several trading strategies have been developed and tested. Results indicate that the proposed approach can give positive signals for what concerns return achievement and reward-to-risk ratio improvement. Nevertheless, due to the high dynamicity of the strategy, as implied by Tactical Asset Allocation hypothesis, transaction costs play a key role in final returns deliverance. All the trading strategies are performed considering the different outcomes of the models.Esta tese tenta desenvolver uma abordagem quantitativa inovadora no desenvolvimento de estratégias de trading baseadas num estilo de investimento de Alocação Tática de Ativos. Este trabalho usa previsões de retornos a partir de algoritmos de aprendizagem supervisionada e estuda se essas previsões conseguem gerar retornos superiores que os de outras estratégias de investimento. Estes modelos conseguem prever os movimentos dos mercados com um nível de precisão aceitável, e ainda sugerem o bet sizing, contribuindo vitalmente para a definição dos pesos de cada ativo. Enquanto a maioria dos estudos alusivos à Alocação Tática de Ativos focaram numa análise histórica, a novidade que esta tese apresenta e a incorporação da previsão de valores futuros no processo de decisão. Neste estudo, foram usados Multi-Layer Perceptron Neural Networks, Support Vector Machines e Random Forests para a previsão dos retornos dos dois principais índices: S&P500 e o Eurostoxx600. Todos os modelos foram usados na sua regressor form para obter o output continuo esperado com o intuito de prever os retornos do próximo dia de ambos os índices. Com base nestas previsões, esta tese criou e estudo várias estratégias de trading. Os resultados dessas estratégias dão sinais positivos na capacidade destes modelos em aumentar os retornos absolutos e os retornos ajustados ao risco. Contudo, devido ao elevado dinamismo deste tipo de estratégias, os custos de transação contribuem significativamente para uma discrepância entre os retornos hipotéticos e os retornos efetivos. Todas as estratégias de trading foram simuladas com base em todos os diferentes outputs dos modelos.Tran, DanVeritati - Repositório Institucional da Universidade Católica PortuguesaLorini, Federico2023-06-26T07:59:34Z2023-02-022023-012023-02-02T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10400.14/41426TID:203277775enginfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-07-12T17:47:00Zoai:repositorio.ucp.pt:10400.14/41426Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T18:34:07.178503Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
Tactical asset allocation : a novel approach via machine learning returns prediction |
title |
Tactical asset allocation : a novel approach via machine learning returns prediction |
spellingShingle |
Tactical asset allocation : a novel approach via machine learning returns prediction Lorini, Federico Tactical asset allocation Machine learning Returns prediction Alocação tactica de activos Previsão dos retornos Domínio/Área Científica::Ciências Sociais::Economia e Gestão |
title_short |
Tactical asset allocation : a novel approach via machine learning returns prediction |
title_full |
Tactical asset allocation : a novel approach via machine learning returns prediction |
title_fullStr |
Tactical asset allocation : a novel approach via machine learning returns prediction |
title_full_unstemmed |
Tactical asset allocation : a novel approach via machine learning returns prediction |
title_sort |
Tactical asset allocation : a novel approach via machine learning returns prediction |
author |
Lorini, Federico |
author_facet |
Lorini, Federico |
author_role |
author |
dc.contributor.none.fl_str_mv |
Tran, Dan Veritati - Repositório Institucional da Universidade Católica Portuguesa |
dc.contributor.author.fl_str_mv |
Lorini, Federico |
dc.subject.por.fl_str_mv |
Tactical asset allocation Machine learning Returns prediction Alocação tactica de activos Previsão dos retornos Domínio/Área Científica::Ciências Sociais::Economia e Gestão |
topic |
Tactical asset allocation Machine learning Returns prediction Alocação tactica de activos Previsão dos retornos Domínio/Área Científica::Ciências Sociais::Economia e Gestão |
description |
The thesis tries to develop a novel quantitative approach for the development of trading strate gies managed through Tactical Asset Allocation investment style. The work employs a returns prediction performed via supervised learning algorithms and studies whether they can help in the deliverance of superior returns compared with the classical buy-and-hold strategies. The models manage to predict the movement of the markets with an acceptable precision level and try to give hints for bet sizing, playing a key role in the weight's definitions. While most of the previously developed Tactical Asset Allocation strategies were performed through a historical analysis. The novelty of the approach comes from the incorporation of these forward-looking predicted values. In thisstudy, Multi-Layer Perceptron Neural Networks, Support Vector Machines, and Random Forests were used to predict the returns of two main market indexes, respectively S&P500 and Eurostoxx600. The models are used in their regressor form to get a continuous output, expected to be the true value of the returns for the following day. Based on this prediction, several trading strategies have been developed and tested. Results indicate that the proposed approach can give positive signals for what concerns return achievement and reward-to-risk ratio improvement. Nevertheless, due to the high dynamicity of the strategy, as implied by Tactical Asset Allocation hypothesis, transaction costs play a key role in final returns deliverance. All the trading strategies are performed considering the different outcomes of the models. |
publishDate |
2023 |
dc.date.none.fl_str_mv |
2023-06-26T07:59:34Z 2023-02-02 2023-01 2023-02-02T00:00:00Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/10400.14/41426 TID:203277775 |
url |
http://hdl.handle.net/10400.14/41426 |
identifier_str_mv |
TID:203277775 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799132067601055744 |