How Helpful are Spatial Effects in Forecasting the Growth of Chinese Provinces?

Abstract

In this paper, we make multi-step forecasts of the annual growth rates of the real gross regional product (GRP) for each of the 31 Chinese provinces simultaneously. Beside the usual panel data models, we use panel models that explicitly account for spatial dependence between the GRP growth rates. In addition, the possibility of spatial effects being different for different groups of provinces (Interior and Coast) is allowed for. We fi nd that both pooling and accounting for spatial effects help substantially to improve the forecast performance compared to the benchmark models estimated for each of the provinces separately. It is also shown that the effect of accounting for spatial dependence is even more pronounced at longer forecasting horizons (the forecast accuracy gain as measured by the root mean squared forecast error is about 8% at the 1-year horizon and exceeds 25% at the 13-and 14-year horizons).

Economic theory and forecasting: lessons from the literature

“…Hernández‐Murillo and Owyang () show that accounting for spatial correlations in regional data can improve forecasts of national employment in the US. Girardin and Kholodilin () also find that allowing for spatial effects when forecasting Chinese output growth results in substantial accuracy improvements. In terms of the usefulness of spatial restrictions for theoretical modelling, I am not aware of existing attempts to do so, but it should not be difficult to embed these types of restrictions into a DSGE model estimated by Bayesian methods, by choosing appropriate priors.…”

Meeting Asia's Infrastructure Needs

“…However, the results suggest that using the GMM-IV estimator has much lower explanatory power as measured by the mean squared error as compared to OLS. Moreover, the GMM-IV estimator has been found to have unsatisfactory and unstable performance in out-of-sample forecasting (Girardin and Kholodilin 2011;Kholodilin, Siliverstovs, and Kooths 2008). Considering that the main purpose of estimating equation 1 is for projecting future stocks of infrastructure, we choose to estimate the model using OLS rather than GMM.…”

Transformation of the Structure of the Gross Regional Product in Siberian Regions

“…maintaining industries) were neglected. Modern China is currently struggling to overcome the consequences of the implementation of this strategy [9]. It is necessary to take into account the fact that this approach had the right to exist under the set geographical and historical conditions but the period of its factual implementation was strongly dependent of the cementing political realia.…”