Curve Registration of Functional Data for Approximate Bayesian Computation

Approximate Bayesian computation is a likelihood-free inference method which relies on comparing model realisations to observed data with informative distance measures. We obtain functional that are not only subject noise along their y axis but also random warping x axis, we refer as the time axis. Conventional distances functions, such L2 distance, under these conditions. The Fisher–Rao metric, previously generalised from space of probability distributions an ideal objective function for aligning one another by assess usefulness alignment metric approximate four examples: two simulation examples, example about passenger flow at international airport, and hydrological modelling. find works well minimise alignment; however, once functions aligned, it necessarily most inference. This means may require distances: parameter