For this talk we focus on quantile additive models (QGAMs), an extension of quantile regression which integrate the flexible additive structure of generalized additive models (GAMs). While QGAMs make fewer assumptions than standard GAMs, they are slower to fit due to the cost of selecting the so-called “learning-rate”. The longer fitting time is particularly problematic when handling large data sets and complex models.

This talk focuses on the development of new Big Data methods for QGAMs (and on their implementation in the qgam package) which much alleviate this issue. The new methods are based on a covariate discretisation approach which leads to a significant decrease in computational time and to much lower memory requirements, while having little or no effect on the accuracy of the predictions. While we will demonstrate the methods on regional solar production modelling, they are useful in a wide range of industrial and scientific applications.