摘要
近年来,黄金和比特币充斥于市场交易,期望实现资产增值。因此,我们计划设计一种交易策略,保证投资风险尽可能小,并评估五年后1000美元的价值。首先对数据进行预处理,由于黄金波动较小而比特币波动较大,所以黄金利用前三十天的价格,比特币利用前二十天的价格,代入GM(1,1)模型,分别预测出黄金和比特币后一天的价格。其次,定义涨跌幅来显示未来一天内涨或跌的幅度与可能性。利用涨跌幅的绝对值与其中位数的绝对值作比较,来决定售出和买入操作。并建立动态规划模型来避免风险,利用蒙特卡洛模拟算法,确定交易时间和交易金额。为证明上述策略为最佳。通过查阅文献固定投资占比,重新计算,得到五年后总价值。通过比较两者的大小,确定策略是最优。改变佣金组合进行灵敏度分析,再次计算,得出每个阶段资金的变化规律以及最后总价值,比较相对增长率来判断敏感性的强弱。
In recent years, gold and bitcoin have flooded the market in hopes of increasing their value. So we plan to design a trading strategy that keeps investment risk as low as possible and estimates the value of $1000 five years from now. Firstly, the data are preprocessed. Since gold fluctuates less while bitcoin fluctuates more, gold takes advantage of the price of the first 30 days, while bitcoin takes advantage of the price of the first 20 days and replaces it with GM(1,1) model to predict the price of gold and bitcoin the next day respectively. Second, a rise or fall is defined to show how much or how likely it is to rise or fall in the next day. The absolute value of the rise or fall is compared to the absolute value of the median to determine the sell and buy operations. A dynamic programming model is established to avoid risks, and Monte Carlo Method is used to determine the transaction time and transaction amount. To prove that the above strategy is the best. By referring to the literature and recalculating the proportion of fixed investment, the total value after five years was obtained. By comparing the sizes of the two, the optimal strategy is determined to change the commission mix for sensitivity analysis, calculate again, get the change rule of funds at each stage and the final total value, compare the relative growth rate to judge the strength of sensitivity.
出处
《应用数学进展》
2022年第4期2187-2203,共17页
Advances in Applied Mathematics