Investment Portfolio Management System Using Machine Learning

Abstract

Abstract: Portfolio management is the concept of determining the proportions of various assets to be held in a portfolio in order to maximize return while minimizing risk exposure. Investment banking and financial management both depend heavily on portfolio optimization. Choosing the greatest feasible combinations of several portfolios to construct an optimal portfolio is an exponentially complex challenge in terms of computation. It's commonly believed that public opinion and financial markets are intertwined. Recently, a variety of machine learning algorithms have been employed to anticipate short-term financial markets with positive outcomes. On the other hand, historical returns don't seem to fit the normal distribution theory. But sentiment analysis performs better when combined with long short-term memory networks and historical data. In this project, we want to use AI/ML to predict portfolio risk and provide insights into how stocks will perform. We will train our model using datasets obtained from the Yahoo Financial API that include historical data from the top 100 companies (NIFTY 100) in the NSE and BSE from 2010 to 2021.