ABSTRACT
Accurate forecasting of the timing and magnitude of macroeconomic recessions caused by unexpected shocks remains an area where both statistical models and judgmental forecasts tend to perform poorly. Inspired by the value-at-risk concept from financial risk management, a growing body of research has been focused on developing a framework to model and quantify macroeconomic risks and estimate the likelihood of adverse macroeconomic outcomes, which has become known as growth-at-risk assessment. The current study proposes an improvement to an established two-step procedure for empirical evaluation of the future growth distribution, which involves directly modelling the parameters of the conditional distribution in one step within an artificial neural network. The proposed procedure is tested on macroeconomic data from four small European open economies covering the coronavirus pandemic lockdown period and the recession related to it. The model achieves a better performance across the four countries compared to the established two-step procedure. © 2022, Bulgarska Akademiya na Naukite. All rights reserved.
ABSTRACT
We use a dynamic panel data model to generate density forecasts for daily active Covid-19 infections for a panel of countries/regions. Our specification that assumes the growth rate of active infections can be represented by autoregressive fluctuations around a downward sloping deterministic trend function with a break. Our fully Bayesian approach allows us to flexibly estimate the cross-sectional distribution of slopes and then implicitly use this distribution as prior to construct Bayes forecasts for the individual time series. We find some evidence that information from locations with an early outbreak can sharpen forecast accuracy for late locations. There is generally a lot of uncertainty about the evolution of active infection, due to parameter and shock uncertainty, in particular before and around the peak of the infection path. Over a one-week horizon, the empirical coverage frequency of our interval forecasts is close to the nominal credible level. Weekly forecasts from our model are published at https://laurayuliu.com/covid19-panel-forecast/.