期刊
JOURNAL OF ECONOMETRICS
卷 195, 期 2, 页码 211-223出版社
ELSEVIER SCIENCE SA
DOI: 10.1016/j.jeconom.2016.09.001
关键词
Spatial correlation; Time-varying parameters; Systemic risk; European debt crisis; Generalized autoregressive scores
资金
- Dutch Science Foundation (NWO) [VICI453-09-005, VENI451-15-022]
- CREATES, Center for Research in Econometric Analysis of Time Series [DNRF78]
- Danish National Research Foundation
- Deutsche Forschungsgemeinschaft [SFB 649]
We extend the well-known static spatial Durbin model by introducing a time-varying spatial dependence parameter. The updating steps for this model are functions of past data and have information theoretic optimality properties. The static parameters are conveniently estimated by maximum likelihood. We establish the theoretical properties of the model and show that the maximum likelihood estimators of the static parameters are consistent and asymptotically normal. Using spatial weights based on cross-border lending data and European sovereign CDS spread data over the period 2009-2014, we find evidence of contagion in terms of high, time-varying spatial spillovers in the perceived credit riskiness of European sovereigns during the sovereign debt crisis. We find a particular downturn in spatial dependence in the second half of 2012 after the outright monetary transactions policy measures taken by the European Central Bank. Earlier non-standard monetary operations by the ECB did not induce such changes. The findings are robust to a wide range of alternative model specifications. (C) 2016 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据