Joint Inference of Misaligned Irregular Time Series with Application to Greenland Ice Core Data

Abstract: Ice cores provide insight into the past climate over many millennia. Due to ice compaction, the raw data for any single core are irregular in time. Multiple cores have different irregularities; jointly these series are misaligned. After processing, such data are made available to researchers as regular time series: a data product. Typically, these cores are independently processed. In this paper, we consider a fast Bayesian method for the joint processing of multiple irregular series. This is shown to be more efficient. Further, our approach permits a realistic modelling of the impact of the multiple sources of uncertainty. The methodology is illustrated with the analysis of a pair of ice cores (GISP2 and GRIP). Our data products, in the form of marginal posterior distributions on an arbitrary temporal grid, are finite Gaussian mixtures. We can also produce sample paths from the joint posterior distribution to study non-linear functionals of interest. More generally, the concept of joint analysis via hierarchical Gaussian process model can be widely extended as the models used can be viewed within the larger context of continuous space-time processes.