Bitcoin chaotic analysis: a price forecasting model proposal

Abstract

Chaos theory describes the behavior of nonlinear dynamic systems and has been used in solving a wide range of problems in economics. The theory of Chaos is based on the assumption that the underlying system is a nonlinear deterministic process. On the other hand, linear models, which are proven to be insufficient in revealing the complexities of economic systems, are not seen as an appropriate modelling approach for analyzing economic data which show chaotic behavior. The focus of this study is two-fold; the first objective is to investigate whether Bitcoin’s non-linear daily price change shows a chaotic behavior. The second objective is to investigate whether the system is suitable for a long-term forecast. The current research investigates both objectives by conducting a prediction model by using a regression model which depends on the embedding size. In order to capture the rich dynamic information hidden in the price changes of Bitcoin, we use the daily closing prices ($) of Bitcoin between the two periods February 2021 and November 2021. The data for such analysis was obtained from marketwach.com.