Estimating Networks: Lasso for Spatial Weights
Pedro Carvalho Loureiro Souza
London School of Economics & Political Science (LSE) - Suntory and Toyota International Centres for Economics and Related Disciplines (STICERD)
March 26, 2012
In applied work, often the spatial neighbouring matrix is assumed known and available to the researcher. This paper provides a method for estimating it within longitudinal data. The salient feature is the ultra-high dimensionality of the problem, which is addressed with an adaptation of the Least Absolute Shrinkage and Selection Operator (Lasso). The main result is that, under identification and sparsity conditions, the estimator is consistent for the true neighbouring matrix, both under L1 and prediction norms.
The proposed model nests a graphical model and suggests several economic applications. Most notably, it provides a framework for estimating networks based on observable cohorts, as opposed to assuming prior knowledge. Finally, Monte Carlo evidence is presented, along with an application for contagion of government bond yields in the wake of the recent European crisis.
Keywords: Networks, LASSO, Spatial regressions, graphical models, longitudinal data
JEL Classification: C01, C13, C31, C33, C45working papers series
Date posted: March 27, 2012 ; Last revised: April 5, 2013
© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo5 in 0.375 seconds