Browsing by Subject "Granger causality"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Developing densely: Estimating the effect of subway growth on New York City land uses
    (Journal of Transport and Land Use, 2011) King, David
    In the early twentieth century, New York City’s population, developed land area, and subway network size all increased dramatically. The rapid expansion of the transit system and land development present intriguing questions as to whether land development led subway growth or if subway expansion was a precursor to real estate development. The research described in this article uses Granger causality models based on parcel-level data to explore the co-development of the subway system and residential and commercial land uses, and attempts to determine whether subway stations were a leading indicator of residential and commercial development or if subway station expansion followed residential and commercial construction. The results of this study suggest that the subway network developed in an orderly fashion and grew densest in areas where there was growth in commercial development. There is no evidence that subway growth preceded residential development throughout the city. These results suggest that subway stations opened in areas already well-served by the system and that network growth often followed residential and commercial development. The subway network acted as an agent of decentralization away from lower Manhattan as routes and stations were sought in areas with established ridership demand.
  • Thumbnail Image
    Item
    Do “Immigrants Increase the Unemployment of US Citizens?” An Empirical Examination of Trump’s Campaign Rhetoric
    (2018) Jensen, Anna
    I analyze the relationship between immigration and the US economy, specifically, the effects on levels of GDP and unemployment. Employing data that spans the time period 1870 to 2015, and using estimation results from a Vector Error Correction Model (VECM) Granger causality/Block exogeneity Wald test (Enders, 2003), I find a long run equilibrium relationship between GDP, unemployment, and immigration inflows that can be specifically described as a bidirectional causality between GDP and immigration, and a unidirectional causality running from immigration to unemployment. Examination of the response of changes in GDP and unemployment levels to a onetime Cholesky innovations (shocks) in immigration, I observe a rise in GDP and a fall in unemployment level. While these observations are relevant for policy making, especially given the current effort to limit legal immigration to the US, I have yet to validate these observations by accounting for the breakdown of the immigrant population into broad geographic regions of their countries of origins, and skill levels, my conclusions should be considered preliminary.
  • Thumbnail Image
    Item
    Learning networks via non-invasive observations
    (2021-05) Dimovska, Mihaela
    Learning the underlying structure of a networked dynamic system from observational data is an important problem in many domains, from climate studies to economics. One of the most well-known approaches to this problem is Granger causality, which relies on the premise that data are sampled at a frequency sufficient to capture the cause-to-effect delays, leading to strictly causal observed dynamics. For such strictly causal systems, it has been shown that Granger causality consistently reconstructs the underlying graph of the network. However, in many domains, such as finance, neuroscience or climate studies, the observed dynamics does not follow the strict causality assumption. Thus, many reconstruction methods that try to deal with non-strictly causal dynamics have been developed in the last decade. These methods, however, tend to put limiting assumptions on the graph structures of networks. Furthermore, as we show in this dissertation, the network reconstruction problem is not well-posed in the non-strictly causal case, as in general, it does not admit a unique solution. Thus, many of the existing methods also do not provide any theoretical guarantees for the learned network. In this work, we develop network reconstruction methods for a large class of networks with non-strictly causal dynamics. We provide theoretical guarantees for the reconstruction, while posing no limitations on the underlying structure. We also address the ill-posedness of the network reconstruction problem. The only required assumption in the novel methods is that at least one strictly causal operator is present in every feedback loop. We also provide orientation rules that can orient some of the non-strictly causal links in the network. We test the methods on several benchmark examples, on random networks via simulations, and we also apply them on real-world datasets that show the effectiveness of the proposed algorithms.