This repository holds the code relevant for the paper entitled GLS Kernel Regression for Network-Structured Data by Edward Antonian, Gareth Peters, Mike Chantler and Hongxuan Yan. In it we we consider the problem of predicting a signal y_t, defined across the N nodes of a fixed graph, over a series of T regularly sampled time points.

In particular, we analyse the case of a set of pollutant monitoruing stations across Californa.

The best place to start is in the notebooks folder. There you will find the following files

This notebook is for downloading and preprocessing the data

This notebook is for visualising the various possible transformations for the data

This notebook is for creating the plots of California

This notebook is for verifying proofs from the paper computationally

This is where the model classes are used and the results from the paper are computed