Robotic process automation for investment modelling

Abstract

The developments in data science, machine learning and artificial intelligence force us to revisit the question “What should be automated and what should be done by humans?” The main objective of our contribution is to apply Robotic Process Automation (RPA) to create a model which identifies risk situations on the model based on the market prices and external data such as M2 Money Stock, Consumer Confidence Index (CCI), Daily Treasury Yield Curve etc., and recommend a proportion of assets in the portfolio. Our goal is to build a model that will beat its benchmark, the S&P 500 index, for that purpose we create a portfolio composed of individual stock titles contained in the S&P 500 index and compare the model rate of return with the real rate of the S&P 500 for the period from 1.1. 2004 to 1.1. 2019. As a result we can show that the cumulative yield of the model beats its benchmark approx. 7 times during the period under review.