No abstract

China implemented a strict lockdown policy to prevent the spread of COVID-19 in the worst-affected regions, including Wuhan and Shanghai. This study aims to investigate impact of these lockdowns on air quality index (AQI) using a deep learning framework. In addition to historical pollutant concentrations and meteorological factors, we incorporate social and spatio-temporal influences in the framework. In particular, spatial autocorrelation (SAC), which combines temporal autocorrelation with spatial correlation, is adopted to reflect the influence of neighbouring cities and historical data. Our deep learning analysis obtained the estimates of the lockdown effects as − 25.88 in Wuhan and − 20.47 in Shanghai. The corresponding prediction errors are reduced by about 47% for Wuhan and by 67% for Shanghai, which enables much more reliable AQI forecasts for both cities.

The transmission of inflation is a widespread occurrence, and managing inflationary pressures is a crucial macroeconomic challenge. Although inflation is a typical macroeconomic variable, its contemporaneous and lagged causal relationships have not been thoroughly investigated, which could result in missing important policy insights. The Bayesian graph vector autoregression (BGVAR) model can identify contemporaneous and lagged causal relationships among economic variables, but it lacks practical research on inflationary inflation. To account for the structural transformation in the inflation transmission process, we propose a Bayesian graph vector autoregressive model with Markov switching (MS-BGVAR), which considers both regime switching and contemporaneous causality among macroeconomic variables. Our study focuses on analyzing the dynamics of inflation transmission relationships among G7 countries under different regimes, as these countries represent developed nations. We use inflation data from 1971-2019, which shows two distinct inflation regimes within the sample period. We conduct simulation experiments to generate moderately dimensional simulated data for both regimes and indicators, demonstrating the theoretical reliability of our model in accurately identifying graph structures. Finally, we apply the proposed model to identify structural breaks and causal transmission relationships in the inflation transmission process of G7 economies, demonstrating that the proposed model has significant economic significance and good explanatory power in the selected target countries.