Portfolio optimization, as a process of reorganizing investment and dispersing risks by considering the established target return and risk tolerance, is widely used in the financial field. This paper analyzes the asset allocation of six kinds of assets. It uses three methods, namely, mean-variance analysis, CAPM, and the FF3F model, to optimize the investment portfolio. At the same time, the proportion of assets in the portfolio should be considered under the conditions of maximizing the return or minimizing the risk at a specific risk level. The results show that in the mean-variance model, the AEP and Gold ETF have the most considerable weight in the combination of maximum Sharpe ratio and minimum variance, respectively. In the FF3F model, AT&T and Gold ETF have the most prominent weight in the combination of maximum Sharpe ratio and minimum variance, while in CAPM, Google and Gold ETF have the most significant weight in the combination of maximum Sharpe ratio and minimum variance respectively. This research may help potential investors interested in power, communication, science and technology, gold, and automobile industries.