ABSTRACT
The COVID-19 pandemic has caused severe public health and economic consequences around the world. It is of great importance to evaluate the impact of the COVID-19 pandemic on the economy, especially the stock market. To this end, we proposed to use several state-of-art sparse principal component analysis (PCA) methods for the stock data of the CSI 300 index from February 1, 2019 to February 1, 2021. To show the influence of the outbreak of the COVID-19 pandemic, we divide this period into two periods, i. e., before and after January 1, 2020. Based on this division, we attempted to extract the principal components and construct portfolio accordingly. The results show that the proportion of principal components representing the market declined after the outbreak. For the constitution in the first two principal components, the important stock sets are substantially different after the outbreak. The stocks from the health care sector start to play an important role in the portfolio of the CSI 300 index after the outbreak. Compared with the CSI 300 index, the first two principal components from the sparse PCA methods can obtain higher returns with a much smaller set of stocks in the portfolio. In conclusion, the outbreak of the COVID-19 pandemic led to changes in both proportion and constitution of the principal component of the stocks in the CSI 300 index. © 2021, Editorial Department of Journal of University of Science and Technology of China. All rights reserved.