ABSTRACT
Corona Virus Disease (COVID-19) has triggered a global disaster by affecting over 200 countries in a short period of time. It has had a significant impact on both social and economic activities all over the world. Panic selling has been prompted by investors' herd behaviour. As a result, stock markets worldwide have plummeted. The market data is nonlinear and chaotic. Predicting the behaviour of the market in current circumstances is a challenging task.. In this work, improved quality of input data and enhanced feature engineering mechanism is adapted to predict the stock market trend amid COVID-19. Here both classical, as well as ensemble regression models are used to investigate the impact of predictors. The models are evaluated using R- squared metric. The findings of this study highlight that the opening price of the market and total positive cases in India significantly impact the closing price of the Nifty50 index. Furthermore, the linear regression model performed better than other models and achieved an R-squared value of 0.999 for both training and test sets. © 2021 IEEE.