Development of smart trading strategy system based on computer dynamic prediction models and algorithm optimization

Abstract

Balancing the relationship between investment risk and return to formulate trading strategies has become an urgent problem in the capital market. In the study, by analyzing the historical price data of gold and bitcoin from 2016 to 2021, the time series method combined with a simple moving average method was introduced to develop price prediction models of gold and bitcoin in the next trading day, accurate predictions of future price for both financial products were completed. Based on this, by comparing the prices of gold and bitcoin two days before and after, six trading types of their asset portfolios were determined, and an optimal decision-making model was established to provide investors with the best daily trading strategy. Furthermore, the sensitivity under different transaction costs was tested by defining the sensitivity of trading strategies to transaction costs. Research informs investors on how to invest for the best returns.