資產配置優化在 AI 理財機器人之應用

科技與時俱進，年輕人紛紛進入理財的世界。而 2018 年金融風暴做為一個分水嶺，使得各家銀行推出了一款能像人類一樣為投資者提供投資組合的建議的機器人也就是理財機器人。理財機器人能夠經由業者所設計的方法為客戶提供合適的投資組合建議，而市場的變化會導致每檔資產的表現不同，因此須適時進行再平衡以維持投資的最大效益。而疫情的出現也對半導體產業產生了不良的影響，而本研究將以費城指數半導體成分股為例，隨機假設四種不同的觀點，探討加入觀點後的 BlackLitterman 模型跟蒙地卡羅方法透過不同的績效指標相比後是否能為投資者提供更好的投資建議。在本文所假設的四種情境下，有兩種情境的 Black-Litterman 模型的投資組合表現相對較好而另外兩種情境則是蒙地卡羅模擬方法的投資組合表現較好。但在四種情境下。本文認為 Black-Litterman 模型能夠加入投資者觀點較能符合投資人的需求，但長期來說蒙地卡羅模擬所得到的投資組合表現可能相對較穩定。

關鍵字

理財機器人；投資組合； Black-Litterman 模型；蒙地卡羅模擬

並列摘要

Technology move with the times. Young people are entering the world of financial management. The 2018 financial crisis served as a watershed moment, which has led to the introduction of a robot that can provide investors with investment portfolio advice like a human being, that is, a robo advisors. Wealth management robots can provide customers with suitable investment portfolio recommendations through the methods designed by the industry. Changes in the market will cause each asset to perform differently, so it is necessary to rebalance in a timely manner to maintain the maximum benefit of investment. The emergence of the epidemic has also had an adverse impact on the semiconductor industry, and this study will take the Philadelphia Index semiconductor stocks as an example, randomly assume four different viewpoints, and discuss the Black-Litterman model and Monte Carlo method after adding viewpoints. Whether it can provide investors with better investment advice after comparing different performance indicators. Among the four scenarios assumed in this paper, the portfolios of the Black-Litterman model in two scenarios performed relatively well and the portfolios from the Monte Carlo simulation method performed better in the other two scenarios. But in four situations. This paper believes that the Black-Litterman model can add investors' opinions to meet the needs of investors, but the performance of the portfolio btained by the Monte Carlo simulation may be relatively stable in the long run.