Two Sides of the Same Coin? Employing Granger Causality Tests in a Time Series Cross-Section Framework

Abstract

In this paper, we introduce a recently developed methodology for assessing the assumption of causal homogeneity in a time series cross-section Granger framework. Following a description of the procedure and the analytical contexts for which it is appropriate, we implement this new approach to examine the transformation of the post-World War II party system in the South. Specifically, we analyze the causal relationship between black mobilization and GOP growth in the region. We find that black mobilization Granger caused Republican growth throughout the South, whereas Republican growth Granger caused black mobilization only in the deep South. We discuss the substantive significance of our results and conclude with guidelines for the appropriate use of this procedure and suggestions for future extensions of the method.

The Weak Relation between Foreign Direct Investment and Corruption: A Theoretical and Econometric Study

“…The main contributors to this topic are Hurlin and Venet (2001), Hurlin (2004a and2004b) and Hurlin and Dumitrescu (2012) and e.g. Hood, Kidd and Morris (2006), as well as several other authors.…”

Exploring causality in trade and air passenger travel relationships: the case of Asia-Pacific, 1980–2010

“…Note that in research that wishes to establish the direction of causality, X and Y can be reversed in the next step, thus leading to the four possibilities of (1) X 'causing' Y (but not the other way round), (2) Y 'causing' X (but not the other way round), (3) no causality, or (4) causality running in both directions. This traditional approach to Granger analysis has been most commonly applied to time series data, where the causal relationship between two characteristics of a single unit is monitored over a certain time period (Hood III et al, 2008).…”

“…and In contrast to traditional Granger causality analysis, TSCS analysis allows monitoring multiple cross-sections in a single analytical framework, making it well suited for analysing larger numbers of cross-sections. Meanwhile, the heterogeneous causality extension of TSCStesting allows for the possibility of dissimilar causation across different cross-sections (see Hurlin and Venet, 2001;Hood III et al, 2008), so that -in addition to the overall causalitythe diversity across cross-sections can be assessed. The method entails the following extension of the traditional Granger model:…”

The Media as a Dual Mediator of the Political Agenda–Setting Effect of Protest. A Longitudinal Study in Six Western European Countries

“…In regular time series, Granger causality is most commonly tested in vector autoregression (VAR) analysis (Vliegenthart 2014), where effects of both x on y and y on x are tested. In the case of pooled time series analysis, a similar logic can be applied (Hood et al 2008). Here, we chose to straightforwardly test the effect of the lagged parliamentary questions in a similar manner as the reversed effect is tested and also investigate whether the effect is mediated by media coverage (see below).…”