Nonparametric Dynamic State Space Modeling of Observed Circular Time Series with Circular Latent States: A Bayesian Perspective

TL;DR

This paper introduces a Bayesian nonparametric state space model for circular time series with circular latent states, using wrapped Gaussian processes and MCMC methods, validated through simulations and real data.

Contribution

It develops a novel Bayesian nonparametric framework for modeling circular time series with unknown, time-varying circular functions and latent circular states.

Findings

Model performs well in simulations and real data applications.

Successfully predicts whale migration directions considering unobserved ocean currents.

Provides a new approach for complex circular time series analysis.

Abstract

Circular time series has received relatively little attention in statistics and modeling complex circular time series using the state space approach is non-existent in the literature. In this article we introduce a flexible Bayesian nonparametric approach to state space modeling of observed circular time series where even the latent states are circular random variables. Crucially, we assume that the forms of both observational and evolutionary functions, both of which are circular in nature, are unknown and time-varying. We model these unknown circular functions by appropriate wrapped Gaussian processes having desirable properties. We develop an effective Markov chain Monte Carlo strategy for implementing our Bayesian model, by judiciously combining Gibbs sampling and Metropolis-Hastings methods. Validation of our ideas with a simulation study and two real bivariate circular time series data sets, where we assume one of the variables to be unobserved, revealed very encouraging performance of our model and methods. We finally analyse a data consisting of directions of whale migration, considering the unobserved ocean current direction as the latent circular process of interest. The results that we obtain are encouraging, and the posterior predictive distribution of the observed process correctly predicts the observed whale movement.