An Analysis of Algorithm-driven Financial Planning Services

Abstract

Since their introduction in 2008, robo-advisors have become more popular among investor with over 100 options now available. Robo-advisors have become popular with novice investors because they don’t require in-depth market knowledge. This might sound like a great choice for individuals that are under prepared for making significant financial decision, but can they really lay back and trust an algorithm to do everything? This thesis seeks to explore the impact of behavioral biases and unfortunate habits with the rise of automated financial planning system, or robo advisors. The main goal of this exploration will be to enhance robo advisors in order to address under prepared investors and lead to higher performance. For this purpose, we will begin with an analysis of the underlying framework behind robo advisors and explore some unfortunate habits that amplify poor performance during investing. We will develop robo-advisory algorithms that automates investments based on individuals' risk tolerance and investment horizon using Modern Portfolio Theory. We will also create 5 portfolios of ETFs ranging from conservative to aggressive for the algorithm to choose from. After creating a robo advisor and its pool of investment portfolios, we will test how different types of investors perform using robo advisor. Based on the performance, we found that the use of a robo-advisor can be beneficial for portfolio management, particularly for those who are prone to behavioral biases such as prospect theory utility, loss aversion bias, and anchoring bias. The passive investor, who relies solely on the robo-advisor, tends to have a smoother trajectory and overall better performance compared to biased investors.