Value-at-risk prediction for the Brazilian stock market: A comparative study between Parametric Method, Feedforward and LSTM Neural Network

Página do item simplificado
dc.contributor.authorReghin D.
dc.contributor.authorLopes F.
dc.date.accessioned2024-03-12T23:52:02Z
dc.date.available2024-03-12T23:52:02Z
dc.date.issued2019
dc.description.abstract© 2019 IEEE.Value-at-Risk (VaR) as a risk quantification mechanism has more than one calculation method, one of which is the parametric method. In the bibliographic study, it was identified that the parametric method is not effective for all market moments, such as those of crisis or abrupt changes in behavior. This study, therefore, seeks to verify whether other methods of calculation are more efficient, such as the use of neural networks. This study compared the VaR calculation using the parametric method against the use of Feedforward neural networks and Long Short-Term Memory (LSTM) recurrent networks. It used the B3 São Paulo Stock Exchange, IBOVESPA, as the index of study. For the parametric method, volatility models such as standard deviation, Exponentially Weighted Moving Average (EWMA) and Generalized Autoregressive Conditional Heteroskedasticity (GARCH) were tested. For neural networks, different layers and amounts of neurons, activation functions, use of different predictors and incorporation of macroeconomic data were explored. The result of the experiment showed that LSTM networks had a better performance when comparing the exception rate generated by the entire model. When analyzing periods of crisis or abrupt changes in behavior, LSTM and Feedforward networks were less efficient in predicting VaR compared to the parametric method.
dc.identifier.doi10.1109/CLEI47609.2019.235095
dc.identifier.urihttps://dspace.mackenzie.br/handle/10899/35207
dc.relation.ispartofProceedings - 2019 45th Latin American Computing Conference, CLEI 2019
dc.rightsAcesso Restrito
dc.subject.otherlanguageArtificial Neural Networks
dc.subject.otherlanguageEWMA
dc.subject.otherlanguageFeedforward Neural Network
dc.subject.otherlanguageGARCH
dc.subject.otherlanguageLSTM
dc.subject.otherlanguageRecurrent Neural Network
dc.subject.otherlanguageValue-at-Risk
dc.titleValue-at-risk prediction for the Brazilian stock market: A comparative study between Parametric Method, Feedforward and LSTM Neural Network
dc.typeArtigo de evento
local.scopus.citations4
local.scopus.eid2-s2.0-85084749881
local.scopus.subjectActivation functions
local.scopus.subjectComparative studies
local.scopus.subjectExponentially weighted moving average
local.scopus.subjectFeed-forward network
local.scopus.subjectGeneralized autoregressive conditional heteroskedasticity
local.scopus.subjectRecurrent networks
local.scopus.subjectRisk quantification
local.scopus.subjectStandard deviation
local.scopus.updated2024-05-01
local.scopus.urlhttps://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85084749881&origin=inward
Arquivos
Coleções
Artigo de evento