Quantifying the economic response to COVID-19 mitigations and death rates via forecasting Purchasing Managers' Indices using Generalised Network Autoregressive models with exogenous variables

TL;DR

This paper introduces advanced GNAR and GNARX models incorporating trade networks and COVID-19 indicators to forecast Purchasing Managers' Indices, demonstrating improved accuracy over traditional models and providing insights into economic impacts of pandemic measures.

Contribution

The paper develops and applies GNARX models with trade networks and COVID-19 data, enhancing forecasting accuracy of economic indicators during the pandemic.

Findings

GNAR models outperform VAR models in forecasting accuracy.

GNARX models further improve predictions by including COVID-19 variables.

Mixed frequency models predict COVID-19 intervention impacts on the UK economy.

Abstract

Knowledge of the current state of economies, how they respond to COVID-19 mitigations and indicators, and what the future might hold for them is important. We use recently-developed generalised network autoregressive (GNAR) models, using trade-determined networks, to model and forecast the Purchasing Managers' Indices for a number of countries. We use networks that link countries where the links themselves, or their weights, are determined by the degree of export trade between the countries. We extend these models to include node-specific time series exogenous variables (GNARX models), using this to incorporate COVID-19 mitigation stringency indices and COVID-19 death rates into our analysis. The highly parsimonious GNAR models considerably outperform vector autoregressive models in terms of mean-squared forecasting error and our GNARX models themselves outperform GNAR ones. Further mixed frequency modelling predicts the extent to which that the UK economy will be affected by harsher, weaker or no interventions.