Predictive Analytics and Portfolio Optimization: A Study on Mutual Fund Asset Allocation and Risk Mitigation

This research involves creating an efficient portfolio construction that aims to guide the retail investors about the significance of the data-driven decision-making using the analytical tool Python, especially for financial securities investments with a focus on mutual funds. A dataset comprising necessary information on nearly 625 mutual fund schemes from the dataset obtained from Kaggle has been utilized for analysis and study. The study focuses on applying the modern portfolio theory for portfolio construction proposed by Markowitz, which is a very popular financial theory, in real-world investment strategy in the case of constructing a mutual fund portfolio with an adjusted risk-return tradeoff. The methodology relies on the construction of a portfolio with modern portfolio theory concepts and predicting the possible outcomes of the portfolio with the Monte Carlo simulation technique by running the codes in Python and constructing two distinct portfolios: one with diversified and lower risk, comprising 15 mutual fund schemes for conservative investors, and another with minimum compromised risk, comprising 5 mutual fund schemes, which achieves a higher return than the previous one. The parameters taken for the choice of selecting the schemes from the data set are based on the renowned ones such as the Sharpe Ratio and Sortino Ratio. The findings reveal that the 15 schemes portfolio returns are in the range of 10% and 12% with risk levels between 1.5% and 2.5%, and the 5 schemes portfolio returns are in the range of 15% and 17% with risk levels between 4% and 4.5%. The optimum weights to be invested in each scheme to achieve maximum return at the lowest possible risk are also mentioned in proportion for both the portfolios. The findings can be interpreted in a way that the construction of a portfolio with rational decisions backed by data is more appropriate in the modern world with the availability of analytical tools such as Python for forecasting and predicting the potential return and constructing the portfolio based on that by minimizing risks with the traditional theories, which can be efficiently and easily used with technology.