Impact of COVID-19 on Square Pharmaceuticals Stock Prices: A Comparative Analysis of Machine Learning Models

Abstract

The aim of this paper is to compare the forecasting performance of different machine learning algorithms in case of the daily stock prices of Square Pharmaceuticals Limited. To ensure the impact of COVID-19 on the stock prices we separated the data into different segments such as pre-COVID period from January 2011 to March 2020, and the COVID period from March 2020 to September 2021 and the whole study period from January 2011 to December 2021. This study compares predicting performance of various machine learning algorithms such as K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Random Forest (RF), Gradient Boosting (GB), and Long Short-Term Memory (LSTM). To ensure a fair comparison of algorithm performance, we implemented the same combination of data splits and time steps consistently across all algorithms, which yielded optimal performance for each model. The empirical findings indicate that the Random Forest model consistently delivered the highest accuracy across all periods, the SVM model showed an unexpected increase in accuracy during COVID period whereas the LSTM model's performance declined. This comprehensive analysis highlights the adaptability and robustness of machine learning models in volatile market conditions, emphasizing their utility in financial forecasting during global disruptions like the COVID-19 pandemic.

International Journal of Statistical Sciences, Vol. 24(2), November, 2024, pp 73-84