Robust Recursive Filtering and Smoothing

TL;DR

This paper introduces a new robust filtering and smoothing method for state-space models that relaxes Gaussian assumptions, is easy to implement, and provides reliable confidence intervals, demonstrated through financial time-series applications.

Contribution

It develops a perturbation-based approach that generalizes existing methods, applicable to diverse models, with simple recursive algorithms and uncertainty quantification.

Findings

Small mean square loss compared to exact methods

Effective in univariate and multivariate models

Successful application to financial data

Abstract

Using a perturbation technique, we derive a new approximate filtering and smoothing methodology generalizing along different directions several existing approaches to robust filtering based on the score and the Hessian matrix of the observation density. The main advantages of the methodology can be summarized as follows: (i) it relaxes the critical assumption of a Gaussian prior distribution for the latent states underlying such approaches; (ii) can be applied to a general class of state-space models including univariate and multivariate location, scale and count data models; (iii) has a very simple structure based on forward-backward recursions similar to the Kalman filter and smoother; (iv) allows a straightforward computation of confidence bands around the state estimates reflecting the combination of parameter and filtering uncertainty. We show through an extensive Monte Carlo study that the mean square loss with respect to exact simulation-based methods is small in a wide range of scenarios. We finally illustrate empirically the application of the methodology to the estimation of stochastic volatility and correlations in financial time-series.