The purpose of this thesis was to empirically test the efficiency of the “new” cryptocurrency market and to further investigate whether using classical, and well-known, statistical arbitrage tools, it was possible to detect and exploit a possible lack of such efficiency. The analysis focused on daily prices, from February 2018 to February 2022, of ten cryptocurrencies. The dataset was handled with two distinct approaches. A full sample analysis, mostly descriptive, and an empirical analysis where the dataset was divided into training set, on which the model parameters were calculated, and test set, where the model was applied and tested to evaluate its results. Through this empirical research it has been possible to apply a "market-neutral" strategy, known as pair trading and based in this study on the concept of cointegration. By introducing different lengths of rolling windows, it has been observed that in some cases the market might not be fully efficient yet.
Lo scopo di questa tesi era quello di testare empiricamente l'efficienza del "nuovo" mercato delle criptovalute e di indagare ulteriormente se utilizzando i classici, e ben noti, strumenti di arbitraggio statistico, fosse possibile rilevare e sfruttare una possibile mancanza di tale efficienza. L'analisi si è concentrata sui prezzi giornalieri, da febbraio 2018 a febbraio 2022, di dieci criptovalute. Il dataset è stato trattato con due approcci distinti. Un'analisi full sample, per lo più descrittiva, e un'analisi empirica dove il dataset è stato diviso in training set, su cui sono stati calcolati i parametri del modello, e test set, dove il modello è stato applicato e testato per valutarne i risultati. Attraverso questa ricerca empirica è stato possibile applicare una strategia "market-neutral", conosciuta come pair trading e basata in questo studio sul concetto di cointegrazione. Introducendo diverse lunghezze di finestre scorrevoli, si è osservato che in alcuni casi il mercato potrebbe non essere ancora completamente efficiente.
Cryptocurrency Market: A Cointegration Based Pair Trading approach
GUALA, MARCO
2020/2021
Abstract
The purpose of this thesis was to empirically test the efficiency of the “new” cryptocurrency market and to further investigate whether using classical, and well-known, statistical arbitrage tools, it was possible to detect and exploit a possible lack of such efficiency. The analysis focused on daily prices, from February 2018 to February 2022, of ten cryptocurrencies. The dataset was handled with two distinct approaches. A full sample analysis, mostly descriptive, and an empirical analysis where the dataset was divided into training set, on which the model parameters were calculated, and test set, where the model was applied and tested to evaluate its results. Through this empirical research it has been possible to apply a "market-neutral" strategy, known as pair trading and based in this study on the concept of cointegration. By introducing different lengths of rolling windows, it has been observed that in some cases the market might not be fully efficient yet.È consentito all'utente scaricare e condividere i documenti disponibili a testo pieno in UNITESI UNIPV nel rispetto della licenza Creative Commons del tipo CC BY NC ND.
Per maggiori informazioni e per verifiche sull'eventuale disponibilità del file scrivere a: unitesi@unipv.it.
https://hdl.handle.net/20.500.14239/1821