Navigation auf uzh.ch

Suche

Department of Informatics Blockchain and Distributed Ledger Technologies

Feature Extraction of Cryptocurrencies’ Price Time Series

Level: MA 
Responsible Person: Yu Zhang
Keywords: Cryptocurrency, time series, feature extraction 

Cryptocurrency is very different from stock though both of them are already listed and traded by investors in the fact that there is no firm business behind to support cryptocurrencies’ price compared to stocks. We already had researched price time series’ classification of cryptocurrency and stock and got an extremely high accuracy rate, which means that the two kinds of time series have very different patterns. So, we wonder what characteristics make the two kinds of time series so different and how we extract the features to explain this result. 

References:

Xiao Z, Xu X, Xing H, Luo S, Dai P, Zhan D. RTFN: A robust temporal feature network for time series classification. Information sciences. 2021 Sep 1;571:65-86.