Title: The network origins of approximate factor models
Authors: Andrew Butters - Indiana University (United States) [presenting]
Scott Brave - Federal Reserve Bank of Chicago (United States)
David Kelley - Federal Reserve Bank of Chicago (United States)
Abstract: The aim is to investigate the large sample properties of the approximate factor model from within a network model framework. After outlining structural network models that exhibit an approximate factor model reduced form, we propose an alternative estimation strategy and provide normal approximation limit theorems for inference. We also provide Monte-Carlo evidence of the efficiency of the proposed estimator in finite samples relative to benchmark estimators (e.g. principal component analysis). Finally, we apply our estimator to an application involving a large cross-section of economic activity indicators.