The assessment of macroeconomic conditions in real time is challenging. Dynamic factor models, which summarize the comovement across many macroeconomic time series as driven by a small number of shocks, have become the workhorse tool for ‘nowcasting’ activity.
This paper develops a novel dynamic factor model that explicitly captures three salient features of modern business cycles: low frequency movements in long-run growth and volatility, lead-lag patterns in the responses of variables to common shocks, and fat-tailed outliers.
Academic researchers from the London Business School, London School of Economics & Political Science, and University of Warwick use real-time unrevised data for the last two decades and cloud computing technology to conduct an out-of-sample evaluation exercise of the model.
The exercise demonstrates the importance of considering these features for forecasting and probability assessment of economic conditions. In an application to the COVID-19 recession, the researchers develop a method to incorporate newly available high frequency data. The use of such alternative data is essential to track the downturn in activity, but a careful econometric specification is just as important.
After the first versions of this paper were circulated, the COVID-19 pandemic and associated recession began to unfold. Researchers use this event as an out-of-sample case study to understand the importance of non-linearities and large shocks in understanding the evolution of economic activity as the crisis unfolds in the United States. In response to the pandemic a number of high-frequency indicators have appeared including credit card transactions, payroll data, or mobility statistics that can give a timely reading of economic conditions but whose available history is sometimes limited to a few months.
The researchers propose a method to incorporate such indicators into their econometric model, and quantify their contribution to the timeliness of the nowcasts. The model is a Bayesian dynamic factor model modified to account for low-frequency variation in the mean and variance of the series, heterogeneity in the impulse responses of variables to common shocks, and fat-tailed observations.
In a concluding paragraph, the paper noted that the researchers have proposed a Bayesian DFM that incorporates low-frequency variation in the mean and variance of the variables, heterogeneous responses to common shocks, outlier observations and fat tails. In a comprehensive evaluation exercise based on fully real-time unrevised data, they have demonstrated that the real-time nowcasting performance is substantially improved across a variety of metrics.
Capturing trends and SV (timevarying long-run growth and stochastic volatility) improves nowcasting performance significantly and heterogeneous dynamics deliver substantial additional improvement. Fat tails successfully capture outlier observations in an automated way and are material to the model’s behavior during the COVID-19 recession. Overall, this paper provides several advances to the nowcasting process.