David Hand’s contribution to the Discussion of ‘New tools for network time series with an application to COVID-19 hospitalisations’ by Nason et al.

Before I comment on the paper, I should remark that I am not an expert on time series analysis.So when I opened the paper on Sunday afternoon I had no great expectation of understanding it, let alone making a contribution to the discussion.But I'd like to commend the authors because, as it turned out, the paper was exceptionally clearly written, and I certainly felt that I understood it.Indeed, I now feel that, at a pinch, I could build a GNAR model myself.However, although I found the description of the GNAR model itself exciting, I felt rather let down by the example.It seemed to obscure rather than reveal the power of such models,