Stock Market and Employment During the COVID-19 Health Crisis

Following the Covid-19 outbreak, stock markets crashed and recovered, and unemployment surged. Our objective was to uncover the effect of sentiments expressed on Twitter (now X) on economic indicators during this volatile period. Gupta et al.'s (2020) dataset included several indicators of sentiments for each of the 63 million coronavirus-related tweets during the health crisis. Stock market indices from Yahoo Finance and unemployment data from Mergent Online were gathered. We applied the LASSO (Least Absolute Shrinkage and Selection Operator) regression to examine the impact of sentiments on economic indicators. Both the Dow Jones and S&P 500 models showed sentiments that were strongly related while the unemployment rate model did not. Over time, even with sentiments expressing lessening joy and increasing anger and fear, the stock market gradually recovered. We present visualizations to depict how unemployment rate over the period was related to sentiments.