Early warnings of systemic risk using one-minute high-frequency data

Abstract

This study uses high-frequency principal component analysis (HF PCA) to extract information from stock prices to monitor and measure systemic risk in the financial system. The empirical analysis carried out in this study using one-minute returns of stocks included in the Russel 3000 index from 2003 to 2021 shows a clear relationship between the size of the realized eigenvalues and systemic increases in financial stress. We also found that realized eigenvectors can trace the role of firms/sectors as potential sources of financial stress in different periods. We measured the transmission of shocks from (to) the financial sector to (from) other sectors and the real economy. This provides a tool for analyzing the spread of this financial instability that could affect the functioning of the financial system to the extent that the real economy is seriously damaged. HF PCA is a risk identification framework that allows policymakers and central banks to detect risks in real-time and address potential threats to financial stability with the most appropriate policy tools.