Optimal trading strategies based on time series analysis

Abstract

Quantitative investment has been widely used in the field of foreign finance, especially the rapid development of international investment in the past decade. And financial activity is an important field of national economic activity. The frequency of financial transactions is an important indicator of the complexity of a country's economy, so it is of great significance to study the optimal investment strategy. This article uses daily price streams from past investments in gold, cash, and bitcoin to determine whether traders should buy, hold, or sell assets in their portfolios. The outlier data were processed by boxplot analysis, and the EM algorithm based on maximum likelihood estimation was used to visualize the case data. The ARIMA model and GARCH model are used to establish the portfolio optimization model and obtain the best portfolio scheme. The time series prediction model is used to conduct specific quantitative analysis on gold and Bitcoin and obtain the investment forecast of the initial $1000 in the future.