The Economic and Social Research Council (ESRC) funded project titled “Cross-sectional Dependence in Panel Data Models” makes an in-depth exploration of cross-sectional dependence in panels. A detailed description of the research project can be found here.

Major themes of interest include:

• Identifying whether cross-sectional dependence is best characterized by spatial (or network) patterns, by unobserved common factors, or by both. In this respect, Chudik, Pesaran and Tosetti (2011) introduce an important distinction between weak and strong forms of cross section dependence and relate these concepts to spatial and factor models.

• The changing nature of cross section dependence over time, such as switches between weak and strong forms. This can shed light on herding behaviour and contagion. With this in mind Pesaran (2012) develops a test of weak cross-sectional dependence for large N and T panels. In the case when cross-sectional dependence proves relatively pervasive, its strength can then be measured as proposed in Bailey, Kapetanios and Pesaran (2012).

• The effect that cross-sectional patterns have on the diffusion behaviour of a variable across time and space. Two important empirical applications that make use of a spatial framework in the context of house prices in the US and UK are Holly, Pesaran and Yamagata (2010) and Holly, Pesaran and Yamagata (2011).

• The nature of cross-sectional dependence relating to the area of unit root testing. Pesaran, Smith and Yamagata (2012) extend the cross-sectionally augmented panel unit root test (CIPS) proposed by Pesaran (2007) and introduce a new panel unit root test based on a simple average of cross-sectionally augmented Sargan-Bhargava statistics (CSB).

Data used in a number of these projects can be found in Links to Data above.

As part of this project we will be hosting a conference on “Cross-sectional Dependence in Panel Data Models” on 30th and 31st May 2013 at the University of Cambridge. The event will be preceded by a “Course on Panel Data Models” on 29th May 2013. It is envisaged that a special issue in the Journal of Applied Econometrics will result from the conference. For further details please refer to Conference and Course links above.

Names and contact details: