Measuring realized interdependencies from volatility and liquidity
Conference
65th ISI World Statistics Congress 2025
Format: IPS Abstract - WSC 2025
Keywords: volatility
Session: IPS 1008 - Modelling Economic and Financial Time Series
Monday 6 October 10:50 a.m. - 12:30 p.m. (Europe/Amsterdam)
Abstract
In this paper we introduce a parametrization for a VAR model capable of capturing the dynamic interdependencies existing across N equities and between M indicators available for each equity; the indicators are recovered from high frequency data and represent realized measures for variance and liquidity. We propose an approach for parameter estimation based on an iterative algorithm that induces sparsity and show its performance with an empirical exercise. Moreover, we provide an economic interpretation for the model parameters in terms of the strength of the links between variables. In addition, building on a model property, allowing us to interpret it as a reduced-form representation of a Structural VAR, we estimate the structural residuals, and provide an analysis of the structural shocks transmission.