Covid-19 Shock: A Bayesian Approach

Abstract

The coronavirus, that started in December 2019, became a pandemic that hit the world economy and had devastating consequences. The spread of the virus suggested preventive measures knowing that no vaccine was available. Therefore, city, district and then country-wide lockdowns were implemented. These variations are introduced as shocks to unemployment, and will be studied in a Vector Autoregressive Framework. The shock to unemployment will be discussed using Bayesian inference. This approach has well-known advantages when studying heavily parameterized models like VARs, as it assigns prior probabilities to the model parameters. These prior beliefs can be updated whenever new information is available. This in turn can help in modelling and forecasting changes that occur following the shock.