Dynamic Network Autoregressive Models for Functional Panel Data

This study proposes a novel functional dynamic network autoregressive framework for analyzing network and dynamic interactions of functional outcomes in panel data settings.

In this framework, an individual's outcome function is influenced by his/her previous outcome and the outcomes of others through a simultaneous equation system.

To estimate the functional parameters of interest, we need to cope with the endogeneity issue arising from these interactions among outcome functions.

We address this issue by developing a novel functional moment-based estimator.

We establish the consistency, convergence rate, and pointwise asymptotic normality of the proposed estimator.

Additionally, we discuss the estimation of marginal effects and functional network impulse responses.

As an empirical illustration, we analyze the demand for a bike-sharing service in the U.S.

The results reveal statistically significant spatial interactions in bike availability, with interaction patterns varying over the time of day.