Forecasting seasonal time series based on fuzzy techniques

Nguyen, Linh; Novák, Vilém

doi:10.1016/j.fss.2018.09.010

Cited by 46 publications

(17 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Nesta seção, a metodologia propostaé avaliada usando a previsão livre ou múltiplos passos a frente, comparando seus resultados com a pesquisa de Nguyen and Novak (2019), que também usam um procedimento semelhante, realizando a previsão de componentes separadamente a priori e, em seguida, agrupando os resultados individuais para prever a série temporal original.À vista disso, as séries temporais reais consideradas nesses experimentos são (Nguyen and Novak, 2019):…”

Section: Resultados Experimentaisunclassified

“…A Tabela 1 a seguir exibe os parâmetros necessários para realização dos experimentos de cada série temporal usando a abordagem proposta por esse trabalho. Destaca-se que: a janela inicial foi considerada w = s; foram extraídas n = 3 componentes não-observáveis de cada série; hé o horizonte tados das NF-TS distribuídas mostraramíndices de erro inferiores em comparação com métodos presentes na pesquisa de Nguyen and Novak (2019), destacando o potencial da metodologia proposta. Para trabalhos futuros, definir um critério para ajustar os parâmetros da metodologia propostaé interessante.…”

Section: Resultados Experimentaisunclassified

“…No trabalho de Joo and Kim (2015), os autores desenvolveram uma abordagem que decompõe a série original em componentes de tendência e outras variações, usando um modelo baseado em filtragem de Wavelets para realizar a previsão separadamente de cada componente. Outro trabalho que segue essa linha está em Nguyen and Novak (2019) Nesse caso, a série temporaĺ e decomposta em componentes cíclicas, de tendência, sazonais e uma componente irregular, sendo que o resultado de previsãoé a combinação da previsão individual de cada componente. As componentes de tendência, cíclicas e sazonais são representadas por modelos baseados em lógica fuzzy, enquanto que a componente irregular usa-se um modelo tradicional Box-Jenkins.…”

Section: Introductionunclassified

See 2 more Smart Citations

Previsão Inteligente de Séries Temporais Baseada em Redes Neuro-Fuzzy Takagi-Sugeno Evolutivas Distribuídas

Júnior

Serra

2019

Anais Do 14º Simpósio Brasileiro De Automação Inteligente

View full text Add to dashboard Cite

This paper proposes an evolving methodology based on Neuro-Fuzzy Takagi-Sugeno networks (NF-TS) distributed for time series forecasting. This proposed approach extracts the unobservable components from time series. Then, the forecast is performed separately for each component, and there is an NF-TS associated with each of these patterns, a fact that highlights the distributed characteristics of proposed approach. The NF-TS are still evolving, i.e., using the data of components to adapt and adjust their structure, where the amount of fuzzy rules of that network can increase or be reduced according to the behavior of components. The method used to extract the components is based on Singular Spectral Analysis technique. The proposed methodology was evaluated and compared with other similar techniques applied for series forecasting, presenting promising results. Resumo: Esse trabalho propõe uma metodologia evolutiva baseada em redes Neuro-Fuzzy Takagi-Sugeno (NF-TS) distribuídas para a previsão de séries temporais. Essa abordagem proposta extrai as componentes não-observáveis a partir de uma série temporal. Em seguida, a previsão de cada componente separadamenteé realizada, sendo que há uma NF-TS associada a cada um desses padrões, fato que destaca a característica distribuída da proposta. As NF-TS são ainda evolutivas, ou seja, usam os dados dessas componentes para adaptar e ajustar sua estrutura, sendo que a quantidade de regras fuzzy dessa rede pode aumentar ou ser reduzida conforme o comportamento das componentes. O método utilizado para extrair as componenteś e baseado na técnica de Análise Espectral Singular. A metodologia proposta foi avaliada e comparada com outras técnicas semelhantes aplicadas para a previsão de séries, apresentando resultados promissores.

show abstract

Section: Resultados Experimentaisunclassified

Section: Introductionunclassified

See 1 more Smart Citation

Previsão Inteligente de Séries Temporais Baseada em Redes Neuro-Fuzzy Takagi-Sugeno Evolutivas Distribuídas

Júnior

Serra

2019

Anais Do 14º Simpósio Brasileiro De Automação Inteligente

View full text Add to dashboard Cite

show abstract

“…This is useful for further analysis of the trend‐cycle (e.g. for forecasting of the trend‐cycle; Nguyen & Novák, 2017; 2019). Unlike the X11‐ARIMA method, the proposed FT‐method can handle any type of time series, not only monthly or quarterly ones.…”

Section: Discussionmentioning

confidence: 99%

“…For instance, future (bull and bear) phases can be obtained by applying the identification technique to the trend extended by its forecasts. It should be emphasized that the trend‐cycle has been successfully predicted using fuzzy techniques (Nguyen & Novák, 2017; 2019). This paper, therefore, focuses on the identification of the past and current phases of the market only, and not on their forecast.…”

Section: Introductionmentioning

confidence: 99%

Trend‐cycle Estimation Using Fuzzy Transform and Its Application for Identifying Bull and Bear Phases in Markets

Nguyen

Novák

Mirshahi

2020

Intell Sys Acc Fin Mgmt

Self Cite

View full text Add to dashboard Cite

Summary This paper is focused on one of the fundamental problems in financial time‐series analysis; namely, the identification of the historical bull and bear phases. We start with the proof that the trend‐cycle can be well estimated using the technique of a higher degree fuzzy transform. Then, we suggest a mathematical definition of the bull and bear phases and provide a novel technique for their identification. As a consequence, the turning points (i.e. the points where the market changes its phase) are detected. We illustrate our methodology on several examples.

show abstract

Inferring efficient operating rules in multireservoir water resource systems: A review

Macian-Sorribes

Pulido-Velázquez

2019

WIREs Water

View full text Add to dashboard Cite

Coordinated and efficient operation of water resource systems becomes essential to deal with growing demands and uncertain resources in water‐stressed regions. System analysis models and tools help address the complexities of multireservoir systems when defining operating rules. This paper reviews the state of the art in developing operating rules for multireservoir water resource systems, focusing on efficient system operation. This review focuses on how optimal operating rules can be derived and represented. Advantages and drawbacks of each approach are discussed. Major approaches to derive optimal operating rules include direct optimization of reservoir operation, embedding conditional operating rules in simulation‐optimization frameworks, and inferring rules from optimization results. Suggestions on which approach to use depend on context. Parametrization–simulation–optimization or rule inference using heuristics are promising approaches. Increased forecasting capabilities will further benefit the use of model predictive control algorithms to improve system operation. This article is categorized under: Engineering Water > Water, Health, and Sanitation Engineering Water > Methods

show abstract

Forecasting seasonal time series based on fuzzy techniques

Cited by 46 publications

References 20 publications

Previsão Inteligente de Séries Temporais Baseada em Redes Neuro-Fuzzy Takagi-Sugeno Evolutivas Distribuídas

Previsão Inteligente de Séries Temporais Baseada em Redes Neuro-Fuzzy Takagi-Sugeno Evolutivas Distribuídas

Trend‐cycle Estimation Using Fuzzy Transform and Its Application for Identifying Bull and Bear Phases in Markets

Inferring efficient operating rules in multireservoir water resource systems: A review

Contact Info

Product

Resources

About