Currently, it is difficult to overestimate the significance of crypto-revolution. Many people pay attention to this market, including financial regulators, who fuss over what they should do with this “free” market, and institutional market majors quietly begin to recruit investment teams to target a cryptomarket. Academic institutions are also starting to explore this sphere actively. For example, LSE launches a 6-week online training course on investing in cryptocurrency at the end of August. By the way, this short course making the theoretical acquaintance with the investing in the cryptocurrency will cost $2300.
Ivy League did not stay on the sidelines, and Yukun Liu and Aleh Tsyvinski, the professors at Yale University, recently contributed to the study of the cryptocurrency market. Their joint research is entitled as “Risks and Returns of Cryptocurrency”. Their voluminous work of 67 pages tries to shed light on the main parameters that can help in forecasting the movement of coins using the example of BTC, ETH, XRP. The main conclusion of this research is that forecasting the movement of crypto assets is “distinct from those of stocks, currencies, and precious metals” (Yukun Liu and Aleh Tsyvinski, 2018). The most reliable and significant factors were the momentum effect and attention of investors.
Regarding the momentum effect, it is a kind of “trend is your friend”: if everything goes up, then it is likely that the next reporting period will show the growth if the market falls, then the fall will continue. Concerning the factor of “investors’ attention”, the authors revealed the correlation between vector of coins’ dynamics and deviation in statistics of request in Google and Twitter. As soon as there was a deviation to the queries with a positive emphasis regarding the coins under study, this led to an increase in prices on the short-term horizon, and when the negative connotation exceeded the norm, the distribution was in favor of a decrease. The study was supported by a representative sample, statistically significant.
In other respects, the connections to the factors identified by professors that could form a certain algorithm for predicting cryptocurrencies, were in fact not statistically significant with rare exceptions. Exposition of this crypto trio in relation to five lead currencies (that were chosen as a possible substitute for their use) showed an insignificant result. The proxy to precious metals, as the store of value, also showed a low level of correlation except for ETH, which had a relatively significant proxy to gold. The assumption of correlation with the cost of mining also did not produce a particularly significant result. ETH was an exception again; it showed some dependence with the dynamics of shares of Advanced Micro Devices Inc. (AMD), one of the leading manufacturers of specialized mining hardware.
As a result of this study, we see that the analysis and forecasting of the cryptocurrencies’ movement is a challenging task. In addition to the already proven and time-tested variety techniques of technical analysis, it is necessary to keep finger on the pulse and if the significant results are confirmed and apply new techniques. It is necessary to monitor breakthroughs in behavioral finance, fundamental analysis in the context of forecasting the movement of cryptocurrencies to improve approaches to analysis and algorithms constantly.
