# Approximate State Space Modelling of Unobserved Fractional Components

## TL;DR

This paper introduces efficient inferential methods for nonstationary multivariate unobserved components models with fractional integration, utilizing ARMA approximations in state space to improve estimation accuracy and computational efficiency.

## Contribution

It presents a novel ARMA-based approximation approach for fractional components models, enhancing estimation accuracy and computational efficiency over traditional truncation methods.

## Key findings

- ARMA approximation outperforms autoregressive/moving average truncation
- Proposed methods show good estimation properties in simulations
- Methods are effective for high-dimensional, complex processes

## Abstract

We propose convenient inferential methods for potentially nonstationary multivariate unobserved components models with fractional integration and cointegration. Based on finite-order ARMA approximations in the state space representation, maximum likelihood estimation can make use of the EM algorithm and related techniques. The approximation outperforms the frequently used autoregressive or moving average truncation, both in terms of computational costs and with respect to approximation quality. Monte Carlo simulations reveal good estimation properties of the proposed methods for processes of different complexity and dimension.