Machine Learning Approach for Finanacial Risk Assessment in Volatile Markets

Abstract

High uncertainty and economic variability in the volatile financial markets. The sectoral portfolio optimization presents an immense challenge to the investor who aims at maximizing the returns and handling risks of the diversified industries. This essay is concerned with the use of Support Vector Machine (SVM) models in risk assessment in volatile markets when it comes to finance. The datasets used to test the performance of SVM in predicting the market risks include S&P 500 financial data. This was subsequently cleaned and normalized and feature engineered by adding volatility values, market trends and sentiment-driven values. The Exploratory Data Analysis (EDA) was carried out to feel the price distributions, correlation of the features and category structures in the data. The Support Vector machine (SVM) model has been used as a prediction stock price movement model since it has been shown to be able to handle nonlinear data series and it works in volatile markets. The data has been divided into training and test sets, and the model has been tested on the basis of R2, MSE and RMSE. The SVM is of high prediction because the values of R2, 0.92, MSE 2.45, RMSE, 1.56. The SVM model was also very accurate when compared to the LSTM, GRU, Prophet and XGBoost models. The research indicates that a business, when combining cyber event-related data with financial forecasts, may find the combination helpful in risk-aware decision-making.