Make the Best Trading Strategy

Zibo Xu

《Frontiers in Economics and Management》 3卷3期 (2022/03) Pp. 581-590

https://doi.org/10.6981/FEM.202203_3(3).0072

摘要

Making trading strategy is a difficult problem that many researches is interested in this field, because it will bring much profit directly. And with the fast development of Bitcoin, the price of it has increase dozens of times. The gold market is relative steady, and the price also rise much. We are asked to develop trading strategy model that can bring significant money. Firstly, we processed the data in SPSS, and make statistical analysis, then we need to forecast the future trend of Bitcoin and gold price. BPNN model has been employed in this paper and the accuracy of BPNN is tested by error measures R2 and MSE, and the result shows high accuracy. Therefore, the models can be applied as the reference to judge whether to buy or not. Then the predicted trend will lead us to buy, hold or sell bitcoins and gold. In financial field, Bollinger Bands strategy (reversion strategy) is a good way for investment decision. So, we combine the BPNN model and Bollinger Band strategy to decide how to operate the bitcoin and gold market. The initial $1000 investment worth $85264.06 on 9/10/2021 using our best trading strategy. At last, there is high relevance between the transaction cost and trading strategy and the correlation coefficient is -0.8425, that means the if the transaction cost rise then the returns from our trading strategy will decline much. And the memorandum has been done.

關鍵字

SPSS ； BPNN ； Bollinger Bands ； Correlation Coefficient

參考文獻

Chen, W. C., et al. "A systematic optimization approach for assembly sequence planning using Taguchi method, DOE, and BPNN." Expert Systems with Applications 37.1(2010):716-726.

Gang, W., et al. "A new approach to intrusion detection using Artificial Neural Networks and fuzzy clustering." Expert Systems with Applications 37.9(2010):6225-6232.

Olsson, U. "Maximum likelihood estimation of the polychoric correlation coefficient." Psychometrika 44.4 (1979):443-460.

Karsoliya, S. "Approximating Number of Hidden layer neurons in Multiple Hidden Layer BPNN Architecture." International Journal of Engineering Trends & Technology (2012).