Dr. Ying Chen is a financial statistician and data scientist. She develops statistical modelling and machine learning methods customized for nonstationary, high frequency and large dimensional complex data such as cryptocurrency, limit order book, and renewable energy. She also works on business intelligence, forecasting, text mining and sentiment analysis, and network analysis. Dr. Chen is Associate Professor in Department of Mathematics and Joint Appointee in Risk Management Institute, National University of Singapore. She also holds Courtesy Appointment in Econ and DSDS.
Webpage: https://blog.nus.edu.sg/matcheny/
| Name of speaker | Chen Ying |
| Schedule | 22 April 2022, 10am – 11.30am |
| Link to register
|
https://nus-sg.zoom.us/meeting/register/tZEucO2pqjosGNPn3FVvmPCZOZrX9T1faGx0 |
| Title of talk | Policy Effectiveness on the Global Covid-19 Pandemic and Unemployment Outcomes: A Large Mixed Frequency Spatial Approach |
| Abstract | We propose a mixed frequency spatial VAR (MF-SVAR) modeling framework to measure the effectiveness of policies conditional on the spillover and diffusion effects of the global pandemic and unemployment. We study the effects of two aspects of policy effectiveness, namely policy start date and policy timeliness, from a spatio-temporal perspective. The spatial panel data contain weekly new case growth rates and monthly unemployment rate changes for 68 countries across six continents at mixed frequencies from January 2020 to August 2021. We find that government policies have a significant impact on the growth of new cases, but only a marginal effect on the change in unemployment rates. A policy’s start date is critical for its effectiveness. In terms of both immediate impact on the near term and total impact over the following four weeks, starting a policy in the 4th week of a month is most effective at reducing the growth of new cases. At the same time, starting in the 2nd or 3rd week is counterproductive for a one-time policy start date. In addition, our estimates suggest that the spillover and diffusion effects are much stronger than a country’s temporal effect during a global pandemic, both for new case growth and changes in unemployment. We also find that new case growth influences changes in unemployment, but not vice versa. Counterfactual experiments provide further evidence of policy effectiveness in various scenarios and also reveal the main risk-vulnerable and risk-spillover countries. This is a joint work with Xiaoyi Han, Yanli Zhu and Yijiong Zhang. The paper is available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4049509 |
