| With the increasing attention paid to bitcoin,the cryptocurrency market represented by bitcoin began to develop rapidly.In 2021,the total market value of cryptocurrency market exceeded US $1 trillion,providing investors with an investment platform with low correlation with traditional financial markets.Due to the same root and source relationship between many mainstream currencies with top market value,the correlation between mainstream cryptocurrencies is very high.After the statistical analysis of the price of mainstream cryptocurrency,it is found that in order to obtain investment benefits in the cryptocurrency market,we can start from the idea of statistical arbitrage and use the momentum factor as the transaction signal to build a cross-sectional investment model.On the basis of reviewing and analyzing the relevant literature of cryptocurrency market research,momentum factor research and statistical arbitrage research,this paper first analyzes how to apply statistical arbitrage model in cryptocurrency market;Then,this paper selects seven mainstream cryptocurrencies with the top market value,and makes a quantitative analysis of their quarterly futures data by using python.It is found that the correlation between mainstream currencies is strong,and the momentum factor difference has the characteristics of mean recovery,which theoretically proves the possibility of statistical arbitrage in the cryptocurrency Market;Then,the specific design of the investment model is carried out.Considering the transaction fee and sliding point,the in sample back test,optimal parameter selection and specific analysis of the investment model are carried out based on the cryptocurrency price data in 2019,and the external model test and analysis are carried out by using the data in 2020.Considering the transaction cost,the annualized rate of return of the investment model is 25%,Sharp ratio is 2.15.The empirical study proves that the cross-sectional statistical arbitrage investment model with momentum factor as the trading signal is effective,which shows that it is feasible to use the idea of statistical arbitrage to trade in the cryptocurrency market.According to the different handling charges,investors can adjust the entry threshold of the investment model and limit the trading frequency;However,in case of extreme unilateral market fluctuations,it is necessary to do a good job in risk management to avoid huge losses.Then the factors affecting the return of the model are analyzed.It is found that the model is less sensitive to the parameters of momentum factor cycle and more sensitive to the parameters of opening threshold and closing threshold.However,when the opening threshold is greater than 0.01 and the closing threshold is greater than-0.01,the return of the model is more stable.Finally,the author summarizes this paper,and puts forward the prospect of the shortcomings of this paper. |
PDF Full Text Request