Partial correlation : Network analysis

The final year project came to us as an opportunity to get involved in a topic which has appeared to be attractive during the learning process of majoring in economics: statistics and its application to the analysis of economic data, i.e. econometrics.Moreover, the combination of econometrics and computer science is a very hot topic nowadays, given the Information Technologies boom in the last decades and the consequent exponential increase in the amount of data collected and stored day by day. Data analysts able to deal with Big Data and to find useful results from it are verydemanded in these days and, according to our understanding, the work they do, although sometimes controversial in terms of ethics, is a clear source of value added both for private corporations and the public sector. For these reasons, the essence of this project is the study of a statistical instrument valid for the analysis of large datasets which is directly related to computer science: Partial Correlation Networks.The structure of the project has been determined by our objectives through the development of it. At first, the characteristics of the studied instrument are explained, from the basic ideas up to the features of the model behind it, with the final goal of presenting SPACE model as a tool for estimating interconnections in between elements in large data sets. Afterwards, an illustrated simulation is performed in order to show the power and efficiency of the model presented. And at last, the model is put into practice by analyzing a relatively large data set of real world data, with the objective of assessing whether the proposed statistical instrument is valid and useful when applied to a real multivariate time series. In short, our main goals are to present the model and evaluate if Partial Correlation Network Analysis is an effective, useful instrument and allows finding valuable results from Big Data.As a result, the findings all along this project suggest the Partial Correlation Estimation by Joint Sparse Regression Models approach presented by Peng et al. (2009) to work well under the assumption of sparsity of data. Moreover, partial correlation networks are shown to be a very valid tool to represent cross-sectional interconnections in between elements in large data sets.The scope of this project is however limited, as there are some sections in which deeper analysis would have been appropriate. Considering intertemporal connections in between elements, the choice of the tuning parameter lambda, or a deeper analysis of the results in the real data application are examples of aspects in which this project could be completed.To sum up, the analyzed statistical tool has been proved to be a very useful instrument to find relationships that connect the elements present in a large data set. And after all, partial correlation networks allow the owner of this set to observe and analyze the existing linkages that could have been omitted otherwise.

Nets: Network estimation for time series

This work proposes novel network analysis techniques for multivariate time series.We define the network of a multivariate time series as a graph where verticesdenote the components of the process and edges denote non zero long run partialcorrelations. We then introduce a two step LASSO procedure, called NETS, toestimate high dimensional sparse Long Run Partial Correlation networks. This approachis based on a VAR approximation of the process and allows to decomposethe long run linkages into the contribution of the dynamic and contemporaneousdependence relations of the system. The large sample properties of the estimatorare analysed and we establish conditions for consistent selection and estimation ofthe non zero long run partial correlations. The methodology is illustrated with anapplication to a panel of U.S. bluechips.

Clustering in complex networks. I. General formalism

We develop a full theoretical approach to clustering in complex networks. A key concept is introduced, the edge multiplicity, that measures the number of triangles passing through an edge. This quantity extends the clustering coefficient in that it involves the properties of two¿and not just one¿vertices. The formalism is completed with the definition of a three-vertex correlation function, which is the fundamental quantity describing the properties of clustered networks. The formalism suggests different metrics that are able to thoroughly characterize transitive relations. A rigorous analysis of several real networks, which makes use of this formalism and the metrics, is also provided. It is also found that clustered networks can be classified into two main groups: the weak and the strong transitivity classes. In the first class, edge multiplicity is small, with triangles being disjoint. In the second class, edge multiplicity is high and so triangles share many edges. As we shall see in the following paper, the class a network belongs to has strong implications in its percolation properties.

Tuning clustering in random networks with arbitrary degree distributions

We present a generator of random networks where both the degree-dependent clustering coefficient and the degree distribution are tunable. Following the same philosophy as in the configuration model, the degree distribution and the clustering coefficient for each class of nodes of degree k are fixed ad hoc and a priori. The algorithm generates corresponding topologies by applying first a closure of triangles and second the classical closure of remaining free stubs. The procedure unveils an universal relation among clustering and degree-degree correlations for all networks, where the level of assortativity establishes an upper limit to the level of clustering. Maximum assortativity ensures no restriction on the decay of the clustering coefficient whereas disassortativity sets a stronger constraint on its behavior. Correlation measures in real networks are seen to observe this structural bound.

Class of correlated random networks with hidden variables

We study a class of models of correlated random networks in which vertices are characterized by hidden variables controlling the establishment of edges between pairs of vertices. We find analytical expressions for the main topological properties of these models as a function of the distribution of hidden variables and the probability of connecting vertices. The expressions obtained are checked by means of numerical simulations in a particular example. The general model is extended to describe a practical algorithm to generate random networks with an a priori specified correlation structure. We also present an extension of the class, to map nonequilibrium growing networks to networks with hidden variables that represent the time at which each vertex was introduced in the system.

Networks of noisy oscillators with correlated degree and frequency dispersion

We investigate how correlations between the diversity of the connectivity of networks and the dynamics at their nodes affect the macroscopic behavior. In particular, we study the synchronization transition of coupled stochastic phase oscillators that represent the node dynamics. Crucially in our work, the variability in the number of connections of the nodes is correlated with the width of the frequency distribution of the oscillators. By numerical simulations on Erdös-Rényi networks, where the frequencies of the oscillators are Gaussian distributed, we make the counterintuitive observation that an increase in the strength of the correlation is accompanied by an increase in the critical coupling strength for the onset of synchronization. We further observe that the critical coupling can solely depend on the average number of connections or even completely lose its dependence on the network connectivity. Only beyond this state, a weighted mean-field approximation breaks down. If noise is present, the correlations have to be stronger to yield similar observations.

Research networks and inventors’ mobility as drivers of innovation evidence from Europe

We investigate the importance of the labour mobility of inventors, as well as the scale, extent and density of their collaborative research networks, for regional innovation outcomes. To do so, we apply a knowledge production function framework at the regional level and include inventors’ networks and their labour mobility as regressors. Our empirical approach takes full account of spatial interactions by estimating a spatial lag model together, where necessary, with a spatial error model. In addition, standard errors are calculated using spatial heteroskedasticity and autocorrelation consistent estimators to ensure their robustness in the presence of spatial error autocorrelation and heteroskedasticity of unknown form. Our results point to the existence of a robust positive correlation between intraregional labour mobility and regional innovation, whilst the relationship with networks is less clear. However, networking across regions positively correlates with a region’s innovation intensity.

Forming efficient networks

In this paper, we suggest a simple sequential mechanism whose subgame perfect equilibria give rise to efficient networks. Moreover, the payoffs received by the agents coincide with their Shapley value in an appropriately defined cooperative game.

Knowledge, networks of cities and growth in regional urban systems

The objective of this paper is to measure the impact of different kinds of knowledge and external economies on urban growth in an intraregional context. The main hypothesis is that knowledge leads to growth, and that this knowledge is related to the existence of agglomeration and network externalities in cities. We develop a three-tage methodology: first, we measure the amount and growth of knowledge in cities using the OCDE (2003) classification and employment data; second, we identify the spatial structure of the area of analysis (networks of cities); third, we combine the Glaeser - Henderson - De Lucio models with spatial econometric specifications in order to contrast the existence of spatially static (agglomeration) and spatially dynamic (network) external economies in an urban growth model. Results suggest that higher growth rates are associated to higher levels of technology and knowledge. The growth of the different kinds of knowledge is related to local and spatial factors (agglomeration and network externalities) and each knowledge intensity shows a particular response to these factors. These results have implications for policy design, since we can forecast and intervene on local knowledge development paths.

Cooperative Transmission and Radio Resources Management for Wireless Networks

Report for the scientific sojourn at the Multimedia Communications Laboratory, University of Texas at Dallas, USA, from September to December 2005. The cooperative transmission has been analyzed taking a broadcast relay channel which assumes a scenario with one source and multiple destinations. Moreover, in order to improve the performance in terms of mutual information, it has been considered that for each destination there is another nearby terminal (called relay) which will help to improve the performance of the destination. This scheme combines different types of channels considered in the information theory, such as the relay channel, broadcast channel and interference channel (if the relays transmit information intended only to its associated destination). In this work, the author has studied the optimal way to encode the signals for the different users, known as capacity region (i.e. related to radio resources management ), of the broadcast relay channel.

The Evolution of roommate networks: a comment on Jackson and Watts JET (2002)

We extend Jackson and Watts's (2002) result on the coincidence of S-stochastically stable and core stable networks from marriage problems to roommate problems. In particular, we show that the existence of a side-optimal core stable network, on which the proof of Jackson and Watts (2002) hinges, is not crucial for their result.

The Difficult and tortuous path of agricultural cooperative networks in spain, 1890-1935

The rural associationism developed from the last decades of the XIX century could be consider as an answer of the agriculturists to the increasing integration of agriculture in the market, and to the effects of the Great Depression. In the case of Spain, the initiatives in this sense arose with certain delay in relation to the countries of Western Europe. The beginning of the Spanish cooperativism is closely bound to the Law of 1906. It granted the agrarian cooperatives with fiscal exemptions and other types of supports to the associates, although the process did not really accelerate until the promulgation of the law regulation in 1908.

Information networks and worker recruitment

This paper studies experimentally how the existence of social information networks affects the ways in which firms recruit new personnel. Through such networks firms learn about prospective employees' performance in previous jobs. Assuming individualistic preferences social networks are predicted not to affect overall labor market behavior, while with social preferences the prediction is that when bilaterally negotiated: (i) wages will be higher and (ii) that workers in jobs with incomplete contracts will respond with higher effort. Our experimental results are consistent with the social preferences view, both for the case of excess demand and excess supply of labor. In particular, the presence of information networks leads to more efficient allocations.

Preliminary study of cooperation in hybrid ad-hoc networks

In this paper, we present a first approach to evolve a cooperative behavior in ad hoc networks. Since wireless nodes are energy constrained, it may not be in the best interest of a node to always accept relay requests. On the other hand, if all nodes decide not to expend energy in relaying, then network throughput will drop dramatically. Both these extreme scenarios are unfavorable to the interests of a user. In this paper we deal with the issue of user cooperation in ad hoc networks by developing the algorithm called Generous Tit-For-Tat. We assume that nodes are rational, i.e., their actions are strictly determined by self-interest, and that each node is associated with a minimum lifetime constraint. Given these lifetime constraints and the assumption of rational behavior, we study the added behavior of the network.

Wealth constraints, skill prices or networks: what determines emigrant selection?

The productive characteristics of migrating individuals, emigrant selection, affect welfare. The empirical estimation of the degree of selection suffers from a lack of complete and nationally representative data. This paper uses a new and better dataset to address both issues: the ENET (Mexican Labor Survey), which identifies emigrants right before they leave and allows a direct comparison to non-migrants. This dataset presents a relevant dichotomy: it shows on average negative selection for Mexican emigrants to the United States for the period 2000-2004 together with positive selection in Mexican emigration out of rural Mexico to the United States in the same period. Three theories that could explain this dichotomy are tested. Whereas higher skill prices in Mexico than in the US are enough to explain negative selection in urban Mexico, its combination with network effects and wealth constraints is required to account for positive selection in rural Mexico.