Pairs trading : cointegration-based methods : applied to the cryptocurrency market

• Autor(a) principal: Carvalho, Daniel da Silva
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.14/36905
• topic: Pairs trading; Cointegration; Cryptocurrency; Time series; Cointegração; Criptomoeda; Séries temporais; Domínio/Área Científica::Ciências Sociais::Economia e Gestão

In statistical arbitrage strategies such as Pairs Trading, the use of cointegration tests has been well-established in the field of econometrics and economics. In this dissertation, the cryptocurrency market was selected to implement, and compare the three best-known types of cointegration tests: the Augmented Dickey-Fuller test, Johansen’s test, and Phillips Peron’s test. Pairs are tested for cointegration over a 3-month and a 6-month window and then traded in a window of the same length. The cryptocurrencies included in the study are 10 cryptocurrencies with the highest market capitalization between January 1st 2020 to July 1st 2021. The performance of each portfolio is compared with their corresponding buy and hold benchmark. Although all portfolios underperformed their buy and hold benchmark, with and without transaction costs, the 6-month trading and testing procedure yielded a higher return compared with the 3-month procedure. Of the three proposed approaches, Augmented Dickey-Fuller test was best at predicting a cointegrated relationship. We can conclude that cointegrated relationships between cryptocurrencies are more likely to hold over longer periods of time and Engle-Granger´s approach employing the Augmented Dickey-Fuller test was the best predictor of cointegration relationships with an excess mean return of 338%.