摘要：Bitcoin has recently been labelled as a “dangerous speculative bubble” by Nobel Prize-winning economists Joseph
Stiglitz and Robert Shiller, as the Bitcoin's market value now exceeds the GDP of over 130 countries. In this study,
the multifractality and efficiency of the Bitcoin price index are tested, using a nonlinear data analysis technique called
the multifractal detrended fluctuation analysis (MF-DFA). In addition, we assess the time-variations in the market
efficiency level through using a rolling-window framework. Our evidence shows that the efficiency of the Bitcoin
market changes over time and this market seems to be more efficient during downward than upward periods. We also
find that Bitcoin is marked by a persistent long memory phenomenon in its short- term components, which could be
interpreted as a possible speculation by investors.
