Motivation

There is an abundance of statistically rigourous methods that allow to deliver reliable answers based on different kinds of data and applications. However, the growing amounts of data and their relative complexity make the use of these methods challenging since they can face numerical issues or can require an excessive (or infinite) amount of computatutional time to output results. Our research is putting forward computational solutions that overcome the computational limitations of existing methods by proposing alternative approaches to implementing these methods or proposing new techniques that benefit from high computational efficiency while preserving statistical soundness.

Publications

• Guerrier, S., Stebler, Y., Skaloud, J. & Victoria-Feser, M.-P., Wavelet-Variance-Based Estimation for Composite Stochastic Processes, Journal of the American Statistical Association (Theory & Methods), 2013. Link Full text

Conference Proceedings