Title: Identifying dominant units using graphical models in panel time series data
Authors: Jan Ditzen - Free University of Bozen-Bolzano (Italy) [presenting]
Francesco Ravazzolo - Free University of Bozen-Bolzano (Italy)
Abstract: An extension of the graphical model is proposed to estimate the covariance matrix using LASSO estimators. The extension allows the use of time-dependent data and aims to identify dominant units in a network. In detail, we propose to loop through the columns of a data matrix which represents the cross-sectional units. Within each loop we obtain a selection of relevant regressors, which inform about the dependence structure of the data. We carry out a Mont Carlo simulation to show that our estimator correctly identifies the dominant units. We illustrate our method by applying it to a dataset of house prices in England.