Abstract
AbstractThis paper provides two specification tests for the system of spatial autoregressive model of order m. We derive the theoretical limit distributions and show in a detailed Monte Carlo simulation study that the tests result in reasonable sized testing procedures with large power. In the empirical application, we analyze Euro Stoxx 50 returns in two different time spans, looking for insights how well models with different specifications of the spatial weighting matrices (local, country, industry and country-industry specific dependencies including interaction effects) fit to the data. The analyzes also demonstrate the ability of the tests to detect inaccurate Value-at-Risk forecasts.
Publisher
Springer Science and Business Media LLC
Subject
Economics and Econometrics,Social Sciences (miscellaneous),Mathematics (miscellaneous),Statistics and Probability
Reference33 articles.
1. Anselin L (1996) Interactive techniques and exploratory spatial data analysis. Regional Research Institute Working Paper, Nr. 9627
2. Arnold M, Stahlberg S, Wied D (2013) Modelling different kinds of spatial dependence in stock returns. Empir Econ 44(2):761–774
3. Asgharian H, Hess W, Liu L (2013) A spatial analysis of international stock market linkages. J Bank Finance 37(12):4738–4754
4. Aue A, Hörmann S, Horvath L, Reimherr M (2009) Break detection in the covariance structure of multivariate time series models. Ann Stat 37(6B):4046–4087
5. Baltagi BH, Song SH, Koh W (2003) Testing panel data regression models with spatial error correlation. J Econom 117(1):123–150
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献