CMStatistics 2017: Start Registration
View Submission - CMStatistics
B0396
Title: Complex oceanic time series observations Authors:  Sofia Olhede - EPFL (Switzerland) [presenting]
Abstract: Oceanic data sets take the form of both Eulerian and Lagrangian observations--observations made at fixed spatial points and via drifting instruments. There are challenges in analyzing instruments that are drifting, especially depending on the velocity of drift, as when the instrument moves, the generating parameters change. Traditional time series models can only encapsulate slow variation in the underlying generative mechanism. However, in many scenarios, this is not a realistic assumption. There seems to be an unavoidable conflict between how rapidly the structured part of the model can change, versus how much we need to average in order to retrieve parameters stably. We introduce a new class of nonstationary time series, and show how efficient and rapid inference is still possible in this scenario, despite the generating mechanism changing quickly. The methods are illustrated on drifter time series, from the global drifter programme. Depending on the latitude of the observations, the underlying generative mechanism of the observed phenomenon is either slowly or rapidly changing, and we show how the newly introduced methodology can resolve both scenarios.