Short-Run Regional Forecasts: Spatial Models through Varying Cross-Sectional and Temporal Dimensions

Abstract

In any economic analysis, regions or municipalities should not be regarded as isolated spatial units, but rather as highly interrelated small open economies. These spatial interrelations must be considered also when the aim is to forecast economic variables. For example, policy makers need accurate forecasts of the unemployment evolution in order to design short-or long-run local welfare policies. These predictions should then consider the spatial interrelations and dynamics of regional unemployment. In addition, a number of papers have demonstrated the improvement in the reliability of long-run forecasts when spatial dependence is accounted for. We estimate a heterogeneouscoefficients dynamic panel model employing a spatial filter in order to account for spatial heterogeneity and/or spatial autocorrelation in both the levels and the dynamics of unemployment, as well as a spatial vector-autoregressive (SVAR) model. We compare the short-run forecasting performance of these methods, and in particular, we carry out a sensitivity analysis in order to investigate if different number and size of the administrative regions influence their relative forecasting performance. We compute short-run unemployment forecasts in two countries with different administrative territorial divisions and data frequency: Switzerland (26 regions, monthly data for 34 years) and Spain (47 regions, quarterly data for 32 years) 2

Does Institutional Quality Matter for Financial Development and Growth? Further Evidence from MENA Countries

“…In this set‐up, we consider the CD, both in the short term and in the long term, by following Holly et al (), who argued that the spatial spillover effect can appear rapidly through various channels, following trade integration, financial integration, neighbourhood, and other networking among the MENA countries. Thus, we address CD issues jointly in the short‐ and long‐term parameters of our models to get a robust result (Mayor and Patuelli ).…”

Regional economic forecasting: state-of-the-art methodology and future challenges

“…For Germany, there are GDP data available on an annual basis for all free cities and counties. Such research activities can further develop the discussion started by Mayor and Patuelli (2012), which mentions the trade-off between the cross section and time dimension. • To date, there is only little evidence of how regional forecast errors are distributed over the business cycle.…”

“…Many authors explicitly mention the necessity of spatial effects to capture, for example, regional spillovers (Baltagi et al, 2014). In this context, the study by Mayor and Patuelli (2012) is worth mentioning. They explicitly examine the trade-off between the cross section and time dimension.…”

Spatial panel data forecasting over different horizons, cross-sectional and temporal dimensions

“…We follow Patuelli et al (2008) and Mayor and Patuelli (2012), and make use of the sign test (ST, Lehmann 1998 …”

“…Both allow to inspect the spatial heterogeneity in the persistence of the regional unemployment rates (and therefore in the shock absorption speed), and to avoid imposing a unique coefficient for all regions . Mayor and Patuelli (2012) analysed the short-run (one-period-ahead) forecasting performance of the above competing spatial models. The SVAR models showed somehow superior performance when the time dimension dominated, consistently with the time-series framework of VAR models.…”