Stock Market Trends During Crisis: A Principal Component Analysis

Abstract

It is undeniable that the COVID-19 pandemic and other such crises noticeably affected the performance of the stock market. The goal of this thesis is to apply Principal Component Analysis to data generated during the pandemic and other crises with the hope of understanding what types of stocks to invest in should similar situations arise in the future. Furthermore, if companies have an understanding of how various crises impact the stock market, appropriate measures could be taken so as to mitigate the harsh effects that have been observed during these times. Using data regarding the 500 stocks in the S&P 500 market acquired from yahoo finance and Anaconda to code using python, I will perform Principal Component Analysis on different financial crises and analyze the results for trends I expect. Additionally, I will perform the same analysis for other years in which no crises were occurring in order to be able to compare and see the full extent of just how much crises change what performs well in the stock market. Armed with knowledge of what characteristics to look for, it is possible to outperform the market should a similar situation occur. If things go according to plan, if another crisis similar to those analyzed were to arise, it would be realistic to mitigate losses and potentially even generate profit during that time. To summarize, rather than analyze the principal components that my code generates, I will analyze the best and worst-performing stocks to see if there are any trends for stocks that tend to “win” during a crisis.