Abstract
Introduction: Advancement in information technology, be it hardware, software or communication technology, over few decades has rapidly impacted almost every field of study. Machine learning tools and techniques are nowadays applied to every field. It has opened the ways for interdisciplinary research by promising effective analyzation and decision-making strategies. COVID-19 has badly affected more than 200 countries within a short span of time. It has drastically affected both daily activities as well as economic activities. Herd behavior of investors has triggered panic selling. As a result, stock markets around the world have plunged down.Methods: In this paper, we analyze the impact of COVID-19 on NSE (National Stock Exchange) index Nifty50. We employ Pearson Correlation and investigate the impact of total confirmed cases and daily cases on Nifty50 closing price. We use various machine learning regression models for predictive analysis viz, linear regression with polynomial terms (quadratic, cubic), Decision Tree Regression and Random Forest Regression. Model performance is measured using MSE (Mean Square Error), RMSE (Root Mean Square Error) and R2 (R Squared) evaluators. Results: Correlation analysis reveals that total confirmed cases and daily cases in both India and the World have negative correlation with Nifty50 closing prices. Moreover, Nifty50 closing prices are more negatively correlated with total confirmed and daily cases in India. Predictive analysis shows that the Random Forest Regression model outperforms all other models. Conclusion: We analyze and predict the impact of COVID-19 on closing price of Nifty50 index. We employ Pearson Correlation and investigate the impact of COVID-19 on Nifty50 closing prices. We use various machine learning regression models to predict the closing price of Nifty50 index. Results reveal that the market volatility is directly proportional to increase in number of COVID-19 cases. Random Forest Regression model has comparatively shown better RMSE and R2 values.