Estimation and Inference in Financial Volatility Networks

This chapter presents a methodological approach for estimating and conducting statistical inference in financial networks of volatility. Starting from economic and financial time series, some networks can be constructed, which quantify the volatility interconnectedness between the agents in a market, to subsequently apply inferential methods to determine the random variables affecting them. Moreover, a review of the existing literature on the latent connectedness of financial volatility and its macroeconomic impact means one step forward to determining the factors affecting it. Thus, this approach allows inferences to be made with the same clarity and simplicity of a standard econometric model but taking connectedness into consideration, which makes it preferable to other techniques. Finally, starting from a consideration of the returns of public debt for a selection of European countries, it is possible to conclude that inflation and financial stress are the main drivers of volatility connectedness and that southern European countries are not necessarily net transmitters of volatility.