Novel network time series models, with applications to advanced modelling and prediction of government flow data sets

Novel network time series models, with applications to advanced modelling and prediction of government flow data sets Cover image

The Principal Investigators on this research project are Professor Matthew Nunes (University of Bath) and Professor Guy Nason (Imperial College London).

The NETF project primarily addresses the current limitations of NARIMA-type models, focusing on methodological challenges in several directions to (a) more adequately reflect reality of the data set (e.g. non-stationary processes and edge processes, multiple network time series); (b) the deceptively simple-sounding challenge of jointly modelling the first- and second-order structure in network time series, permitting better simultaneous estimation of both, building on the ‘multiscale transforms on a graph’ work; (c) incorporate extra, new and valuable information that previous models fail to account. The impact is explicit and will be progressed in our collaborations with ONS and BT.  Amandine Pierrot is the PDRA working on this project, bringing her expertise in industrial forecasting.

Our Research

See other projects

02

Autoregressive network models with stylized features of network data

We will propose several dynamic network models based on a simple AR(1) network framework. The setting depicts the dynamic changes for network edges e

Learn more Learn more button
03

Modelling and forecasting dynamic networks via their edges

Networks that arise in fields such as biology or energy present features that challenge established modelling setups since the target function may nat

Learn more Learn more button
Email subscription

Stay up to date with our events